NOVEL APPLICATIONS OF COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAPHY TIME-OF-FLIGHT MASS SPECTROMETRY

by

Amy L. Payeur

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Chemistry) in The University of Michigan 2011

Doctoral Committee: Professor Robert T. Kennedy, Co-Chair Professor Richard D. Sacks, Co-Chair (Deceased) Professor Mark E. Meyerhoff Emeritus Professor Philip A. Meyers Professor Michael D. Morris Associate Professor Kristina I. Hkansson

 Amy L. Payeur 2011

To Mom and Dad, with love.

ii

Acknowledgements

If nothing else, my graduate school experience has been unique, and has certainly made me not only a stronger scientist but also a much stronger person. Its hard to really thank everyone who had a hand in making my time here everything that it was, but, I am definitely going to try. Id first like to thank my advisor, Dr. Robert T. Kennedy, for all of his guidance and support. The stronger scientist part is mostly his doing. But I also owe him a thank you for taking me on in my second year and making my transition from the Sacks Lab to the Kennedy Lab as easy as he possibly could. I would also like to thank my committee Dr. Mark Meyerhoff, Dr. Michael Morris and Dr. Kristina Hakansson for all of their help as well. A very special thank you goes to my cognate member Dr. Philip Meyers for not only being a great geo advisor but for also being incredibly supportive of me both personally and professionally. Thank you to the Kennedy Lab members both past and present; many of you have been both fantastic lab mates and wonderful friends. Dr. Kendra Reid Evans, Maura Perry, Gwen Anderson, Dr. Omar Mabrouk, Dr. Claire Chisolm, Dr. Hernan Fuentes and Dr. Anna Clark: thank you for always believing in me and always knowing when I needed a Ben & Jerrys break. This accomplishment would not have been possible without the encouragement, support and understanding of my friends Dr. John Henssler, Dr. Nick Deprez, Dr. Jon

iii

Mortison, Dr. Cornelius Kristalyn, Dr. Max Bailor, Dr. Caleb Bates, Katrina Lexa, Matt and Ahleah Rohr Daniel, Dr. Chris Avery, Brad and Sarah Grincewicz, Katie Frey, Diedre Murch, Dr. Andrew Higgs, and Dr. Antek Wong-Foy. Knowing that you were all (and always will be) in my corner made each day of graduate school just a little bit easier, Im not sure what I would have done without each of you. Kristin Bonomo and Stephanie Perry: thank you for being two of my biggest fans not only during my time here at Michigan but back in the day at Union as well. Thank you to everyone at Leco and Restek for their technical guidance and friendship over the years, especially, Joe Binkley, John Heim, Todd Barton, Chris Immoos, Lucas Smith, Frank Dorman and Jack Cochran. Thank you to all the Sacks Lab alumni who made my first two years of graduate school absolutely amazing. I have never met a group of people who epitomized the adage Work Hard, Play Hard as well as you all. Dr. Joshua Whiting, Dr. Mark Libardoni, Dr. Randy Lambertus, Dr. Cory Fix, Dr. Peter Stevens, Dr. Shaelah Reidy, Dr. Shai Kendler, Dr. Juan Sanchez and Meg Ziegler, thank you, for your continued friendship and support; The Chromies will always hold a very special place in my heart. Dr. Megan McGuigan, it always feels like thank you is never enough. The role you have played in my life as a mentor, a colleague, and most importantly a friend is absolutely invaluable and I will never be able to truly thank you for everything that you do. Dr. Richard Sacks, where do I begin? Thank you for your contagious enthusiasm and love of science. Thank you for your encouragement, your understanding and for being an amazing mentor. Although my time with you was way too short, you have left

iv

an impression with me forever and I will always be grateful that I was able to work with you. I think of you every time a thunderstorm rolls through; thanks for checking in, I needed the extra push. Thank you to my amazing family especially Mom, Dad, Nick, and Mimi, I definitely would not be here if it werent for your love, your support and your ability to at least pretend to understand why I was still in school all this time . Thank you to my Memere and Pepere, who are not here to physically see this day, but are no doubt looking down, smiling and extremely proud of what their granddaughter has accomplished. Finally, thank you, Dr. William Porter. Words cannot truly express how blessed I feel to have had someone who believes in me the way that you do as my partner through most of this journey; I cant wait for our next trip together.

TABLE OF CONTENTS

DEDICATIONS ................................................................................................................ ii ACKNOWLEDGEMENTS ............................................................................................ iii LIST OF FIGURES ......................................................................................................... ix LIST OF TABLES ......................................................................................................... xiii LIST OF APPENDICES .................................................................................................xv CHAPTER 1. INTRODUCTION .................................................................................... 1 Gas Chromatography Background ..................................................................... 1 Comprehensive Two-Dimensional Gas Chromatography Background .......... 6 Peak Capacity in GC GC .................................................................................11 Dissertation Overview .........................................................................................13 References ............................................................................................................ 15 CHAPTER 2. METABOLITE PROFILING AND METABOLOMIC ANALYSIS OF INS-1 CELLS USING COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAPHY TIME-OF-FLIGHT MASS SPECTROMETRY .............. 16 Introduction ..........................................................................................................16 Experimental ........................................................................................................22 Results ...................................................................................................................26 Discussion..............................................................................................................46

vi

Conclusions ...........................................................................................................65 References .............................................................................................................66 CHAPTER 3. ANALYSIS OF LIPID COMPOSISTION IN INS-1 CELLS VIA COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAHY TIME-OF-FLIGHT MASS SPECTROMETRY ..........................................................68 Introduction ..........................................................................................................68 Experimental ........................................................................................................70 Results and Discussion.........................................................................................73 Conclusions ...........................................................................................................85 References .............................................................................................................87 CHAPTER 4. PILOT STUDY OF WHOLE SEDMENT PYROLYSIS COMPREHENSIVE TWO-DIMENSIONAL GAS CROMATOGRAPHY TIMEOF-FLIGHT MASS SPECTROMETRY (PY-GC GC-TOFMS) ON A MEDITERRANEAN SAPROPEL SYQUENCE ..........................................................88 Introduction ..........................................................................................................88 Experimental ........................................................................................................92 Results and Discussion.........................................................................................97 Conclusions .........................................................................................................114 References ...........................................................................................................116 CHAPTER 5. SUMMARY AND FUTURE WORK ..................................................118 Summary .............................................................................................................118 Future Work .......................................................................................................120 References ...........................................................................................................128

vii

APPENDICES ................................................................................................................129

viii

LIST OF FIGURES

Figure 1.1:

Golay plot for 0.25 mm i.d. thin-film columns of various lengths using helium as carrier gas GC GC instrument schematic showing C1 (column 1) connected in series by a low dead volume connection through the thermal modulator to C2 (column 2) which continues through a transfer line to the detector Theorectical demonstration of peak capacity achieved with orthogonal first and second dimension columns Schematic of unmodulated (A) and modulated (B) peaks in GC GC Schematic of data processing performed in GC GC showing chromatograms being chopped and merged to display contour plots based on modulation period GC GC chromatogram of fatty acid methyl esters emphasizing the elution of homologous series along an arc (A) and clustered elution of compound classes (B) Schematic of glucose stimulated insulin secretion (GSIS) Evidence of KATP-channel independent pathway KATP channel-dependent (left) and KATP channel-independent (arrows on right) glucose signaling pathways in the -cell are shown Schematic of glycolysis, the pentose phosphate shunt, and the citric acid (TCA) cycle Total ion chromatogram of the commercially available amino acid standard Total ion chromatogram of glycolysis and TCA standards ix

Figure 1.2:

Figure 1.3:

Figure 1.4:

Figure 1.5:

Figure 1.6:

10

Figure 2.1: Figure 2.2: Figure 2.3:

18 19 20

Figure 2.4:

21

Figure 2.5:

27

Figure 2.6:

27

Figure 2.7: Figure 2.8:

Calibration curve for proline Glucose dose response curve where maximal insulin release occurs at ~10 mM glucose Total ion chromatogram of a 17 mM INS-1 cell extract (top) Fisher Ratio plots for randomized 3 mM glucose groups (A), 3 mM glucose compared to 7 mM glucose (B), 3 mM glucose compared to 17 mM glucose (C) and 7 mM glucose compared to 17 mM glucose (D) Zoomed in Fisher Ratio plots for 3 mM versus 3 mM glucose (left) and 3 mM versus 17 mM glucose (right) with red line indicating the 1064 threshold Histograms of Fisher Ratios for 3 mM vs 3 mM glucose (top left) and 3 mM vs 17 mM glucose (top right) Flow-chart summarizing the process used for determining analytes of interests in the 7 mM to 17 mM glucose data set after Fisher Ratio analysis Pathway map for 3 mM to 7 mM glucose data set obtained from Metscape Pathway map for 3 mM glucose to 17 mM glucose obtained from Metscape Pathway map for 7 mM glucose to 17 mM glucose obtained from Metscape Effect of glucose on detectable glycolysis analytes Effect of glucose on detectable TCA and pentose phosphate shunt (R5P) analytes Effect of glucose on detectable amino acids Linoleic acid metabolism pathway as obtained by KEGG database Plot of the average peak area of arachidonic acid at 3, 7 and 17 mM glucose indicating the substantial increase of AA at 17 mM glucose x

28 29

Figure 2.9: Figure 2.10:

31 35

Figure 2.11

36

Figure 2.12

37

Figure 2.13:

39

Figure 2.14:

41

Figure 2.15:

42

Figure 2.16:

43

Figure 2.17: Figure 2.18:

47 47

Figure 2.19: Figure 2.20:

48 50

Figure 2.21:

51

Figure 2.22:

Arachidonic acid metabolism pathway as obtained from KEGG database Butanoate metabolism pathway obtained from KEGG database Glycerophospholipid metabolism pathway obtained from KEGG database Glycosphingolipid metabolism pathway as obtained from KEGG database Vitamin B3 (nicotinate and nicotinamide) metabolism as obtained from the KEGG database Vitamin B5-CoA biosynthesis from pantothenate pathway as obtained from the KEGG database The urea cycle and metabolism of arginine and proline as obtained from the KEGG database Tyrosine metabolism pathway as obtained from KEGG database GC GC chromatogram of neat 37 component FAMEs mix where n is the number of double bonds GC GC total ion chromatogram (TIC) of a representative INS-1 cell extract (top) Calibration curve for myristic acid Average area of palmitic acid (C16:0), stearic acid (C18:0), eicosenoic acid (C20:1), arachidonic acid (C20:4), behenic acid (C22:0) and erucic acid (C22:1) at 0 mM, 0.5 mM, 10 mM, and 20 mM glucose GCMS TIC chromatogram of black shale containing two UCMs. (B) GC GCMS total ion chromatogram of the same sample with the labeled n-alkanes (black circles), mono-, bi-, tri-, tetra- (steranes), and pentacyclic (hopanes) GC GC chromatogram of an EPA method limestone extraction (top) and pyrolysis GC GC chromatrogram of an identical limestone samples (bottom)

52

Figure 2.23: Figure 2.24:

55 57

Figure 2.25:

58

Figure 2.26:

60

Figure 2.27:

61

Figure 2.28:

63

Figure 2.29:

64

Figure 3.1:

74

Figure 3.2:

76

Figure 3.3: Figure 3.4:

78 83

Figure 4.1:

90

Figure 4.2:

91

xi

Figure 4.3:

Location of ODP Site 974 in the Tyrrhenian Basin of the Mediterranean Sea Photo of core used for pyrolysis GC GC analysis GC GC total ion chromatogram (TIC) of sapropel interval 119-120 cm (A) GC GC total ion chromatogram (TIC) of non-sapropel interval 112-113 cm Fatty acid methyl esters in INS-1 cells combined with an isotopically labeled standard GC GC chromatogram using variable modulation

93

Figure 4.4: Figure 4.5:

94 99

Figure 4.6:

100

Figure 5.1:

123

Figure 5.2:

127

xii

LIST OF TABLES

Table 2.1: Table 2.2: Table 2.3:

List of target metabolites Technical relative standard deviations for metabolite profiling analysis Biological relative standard deviations for metabolite profiling analysis Active metabolic pathways, as indicated by Metscape, for metabolomic analysis List of isolated analytes as indicated by Metscape analysis Average technical variability presented as relative standard deviations (RSDs) at 0 mM, 0.5 mM, 10 mM and 20 mM glucose Biological variability presented as relative standard deviations (RSDs) at 0 mM, 0.5 mM, 10 mM and 20 mM glucose Determination of fatty acids in INS-1 cells incubated for 60 min at different glucose concentrations Samples of ODP Site 974 (Tyrrhenian Basin) insolation cycle 94 sapropel sequence used for pyrolysis GCGC-ToFMS analyses Alkanes (x) and branched alkanes () identified in the respective intervals Alk-1-enes and alk-2-enes identified in respective intervals. X indicates a visible peak but the absence of a software peak marker Furans, thipohenes, and pyrroles, identified in respective intervals. ND = not detected, X = detected Naphthalenes and phenols identified in respective intervals Benzene and indane isomers identified in respective intervals xiii

30 32 33

Table 2.4:

44

Table 2.5: Table 3.1:

45 79

Table 3.2:

80

Table 3.3:

82

Table 4.1:

95

Table 4.2:

103

Table 4.3:

104

Table 4.4:

106

Table 4.5: Table 4.6:

108 109

Table 4.7:

Methyl ketones identified in respective sapropel samples

112

xiv

LIST OF APPENDICES

Appendix A: List of Peaks Identified as changing by Fisher Ratio Analysis from 3 mM glucose to 7 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas Appendix B: List of Peaks Identified as changing by Fisher Ratio Analysis from 3 mM glucose to 17 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas

129

133

Appendix C: List of Peaks Identified as changing by Fisher Ratio Analysis from 7 mM glucose to 17 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas Appendix D: List of Pathways Identified by Metscapse Analysis from 3 mM Glucose to 7 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas Appendix E: List of Pathways Identified by Metscapse Analysis from 3 mM Glucose to 17 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas List of Pathways Identified by Metscapse Analysis from 7 mM Glucose to 17 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas

137

141

152

Appendix F:

164

Appendix G: Location of Raw and Processed Data Files for Chapters 2 & 3

176

xv

Chapter 1 INTRODUCTION Gas Chromatography Background Gas Chromatography (GC) is the most widely used analytical technique for the separation of volatile and semi-volatile organic compounds. The popularity of this technique can be attributed to the ease of use, the relatively low cost of instrumentation, the wide variety of detectors available, and the possibility of rapid, high resolution separations. GC has been used for numerous applications including the separation of essential oils1, 2, environmental studies3, forensics4, 5 and in clinical research 6. The separation produced by a chromatographic system is influenced by many factors. In capillary GC, these include column length, inner diameter (i.d.), stationary phase film thickness, carrier gas type, flow rate, detectors, and inlets. In order to more directly compare the general separation performance from system to system, or column to column, a number of metrics have been developed. One of the most common metrics used to compare systems is efficiency. Efficiency is described by the length of column required to obtain the equivalent separation that would occur under equilibrium conditions for specified values of distribution ratio (K) and phase volume ratio (Vr).7 This length is called the height equivalent to a theoretical plate (H).

H is best described using the kinetic model which was introduced by Golay in 19588 for open tubular columns and takes into consideration the rates of various processes that contribute to band dispersion, Equation 1.1. (1.1)

B is the longitudinal diffusion term, f1 is the Golay-Giddings gas compression correction factor, f2 is the Martin-James gas compression factor, C g contains the contributions from the resistance to mass transport in the mobile phase and band broadening due to parabolic laminar flow effects, Cs is the resistance to mass transport in the stationary phase, and avg is the average carrier gas velocity. The gas compression factors f1 and f2 are described in Equations 1.2 and 1.3

(1.2)

(1.3) where P is the ratio of inlet to outlet pressure. The longitudinal diffusion term B, derived from Einsteins equation for one -dimensional diffusion, describes peak broadening as a consequence of the residence time of the solute within the column and the nature of the carrier gas.9 This term is defined in Equation 1.4

(1.4) where Dg is the binary diffusion coefficient of the analyte in the carrier gas. The effect of this term becomes significant only at low carrier gas velocities; because it is inversely 2

proportional to avg . B is only a minor contributor to band broadening at high average carrier gas velocities. Resistance to mass transport, Cs and Cg, are non-equilibrium band broadening caused by the finite time required for a solute molecule to move from one of the phases to the other phase while they are carried through the column by carrier gas flow. The Cg term also includes band broadening caused by Taylor dispersion or parabolic laminar flow (maximum flow at column center, minimal flow at column walls) effects which cause band broadening due to analytes in these regions travelling at different local carrier gas velocities. Equations 1.5 and 1.6 describe the Cg and Cs terms of the Golay equation (1.5)

(1.6)

where k is the capacity factor, r is the inner radius of the column, df is the stationary phase thickness and Ds is the binary diffusion coefficient for the analyte and stationary phase. The radius of commercially available columns is usually three orders of magnitude greater than the film thickness so the Cs is often overwhelmed by the Cg term and therefore neglected. Golay plots (plate height vs. carrier gas velocity) can be used to visualize and evaluate the effects that chromatographic parameters have on separation efficiency. For example, Figure 1.19 shows efficiency increasing substantially with decreasing column length at high carrier gas velocities for 0.25 mm i.d. columns, using helium as a carrier gas at 50 C and with k and Dg values of 5.0 and 0.4 cm2/s, respectively. 3

Figure 1.1 Golay plot for 0.25 mm i.d. thin-film columns of various lengths using helium as carrier gas. A binary diffusion coefficient of 0.4 cm2/s and a retention factor of 5.0 are assumed. 9

However, column resolving power, another metric to be considered, decreases steadily with decreasing column length. The number of theoretical plates, N, is a measure of the width of sample bands as they elute from the column. N is defined in Equation 1.7 (1.7)

where L is the length of the capillary column. Note that high efficiency does not directly correlate to high resolving power and it is possible for a long, low efficiency column to have better resolving power than a short, high efficiency column.

Another important method for evaluating a GC separation is peak capacity. Peak capacity is a measure of how many completely resolved peaks can fit within the time of the chromatogram at a defined resolution, or the ratio of peak separation to average base peak width, and is given by Equation 1.8 (1.8) where Rs is the user-defined resolution, tRL is the retention time of the last eluting component, and tM is the time it takes for an unretained analyte to reach the detector, also known as the hold-up time. Based on this equation, a 30 m, 0.25 mm i.d. capillary with 4,000 plates per meter, a tM of 1 min., a of 50 cm/s and a runtime of 30 min would have a peak capacity of 250 peaks. However, Equation 1.8 assumes that the mixture components elute with perfect spacing, thus obtaining useful information for the entire time window of the chromatogram. In real samples, this perfectly spaced elution does not occur; instead, peaks tend to be randomly distributed in the chromatogram so that the probability of peak overlap is high in complex mixtures. It is typical that the peak capacity requirement is greater than the number of components in a mixture if all analytes are going to be resolved. Statistical analysis has shown that the required peak capacity may be nearly 20 times the number of peaks in the chromatogram in order to separate completely about 90 % of the peaks.10,
11

For example, to resolve 90 out of 100

components a peak capacity of 1910 would be required. 11 Many current interests in the area of chromatography focus on extremely complex samples such as petrochemicals,12, contain >>1000
13

fragrances,14 and metabolomics,15,

16

that can

species and therefore high peak capacity is essential. In 1991, a

remarkable advancement in peak capacity was made by the late John Phillips with the introduction of comprehensive two-dimensional gas chromatography (GC GC).17 Comprehensive Two-Dimensional Gas Chromatography Background Over the past two decades, GC GC has developed into a popular method for the separation of complex mixtures in research laboratories. GC GC has been used to analyze biological, environmental, food, forensics, pharmaceutical and fragrance samples,18 and the growing popularity of this technique is indicated by the nearly seven times increase in the number of publications per year since 2000. Figure 1.2 shows a schematic of a typical GC GC instrument.

Figure 1.2 GC GC instrument schematic showing C1 (column 1) connected in series by a low dead volume connection through the thermal modulator to C2 (column 2) which continues through a transfer line to the detector. C1 and C2 are housed in independently temperature programmed ovens.

The key to this instrument is the placement of two columns in series with a modulator interface between them. The modulator provides the second, relatively short column with smaller subsets of the original matrix eluting from the relatively long, primary column, and the second column generates a series of high-speed separations.19,
20

The two

columns separate analytes based on different molecular properties. The first column is 6

usually non-polar, separating analytes primarily based on volatility and the second column typically has a polar stationary phase that separates components by polarity. Ideally, the two dimensions in a GC GC separation would operate statistically independent and the entire two dimensional plane of the chromatogram would be available for peak separation.21 This is often referred to as an orthogonal separation and is illustrated in Figure 1.3 where (a) demonstrates the separation space available for a one dimensional separation, (b) represents what would be obtained from performing a separation on two columns connected in series with identical stationary phase chemistries and (c) shows the separation space available in two-dimensional chromatography when orthogonal columns are employed.

Figure 1.3 Theoretical demonstration of peak capacity achieved with orthogonal first and second dimension columns. Peak capacity possible in one-dimension (a); peak capacity possible with two columns of identical stationary phase chemistries connected in series (b); theoretical peak capacity in and orthogonal GC GC separation (c).

The modulator is used to trap and focus a portion of a band eluting from the first column and then periodically inject it as a narrower, more concentrated band into the second column. With the dual-stage thermal modulator commercially available through Leco Corporation (St. Joseph, MI), this is accomplished using a series of liquid-nitrogencooled nitrogen jets and hot air jets. Valve modulators, resistively heated modulators, cryogenic modulators, and additional jet based modulators have also been used. Each 7

type of modulator has advantages and disadvantages including temperature limitations, robustness, portability and consumption of cryogens. 22 Although most current GC GC work is focused on applications, modulators continue to be an area of active development. Besides peak capacity increase, GC GC also can improve sensitivity because of the effect of the modulator at the end of the first column. Figure 1.4 shows a conceptual comparison between an unmodulated peak (A) and a modulated peak (B); the area of the unmodulated peak is equal to the sum of the area of the modulated peak.23; however, the intensity of the narrow modulated peak slices is 10-50 times the height of the unmodulated peak. This greatly increased peak height, caused by the focusing action of the thermal modulator, significantly increases detectability, thus making this technique well-suited for trace level analytes that would not be detected in one-dimensional GC.

Figure 1.4 Schematic of unmodulated (A) and modulated (B) peaks in GC GC. Adapted from reference23

The output of the GC GC is a string of very fast separations that are in 2-20 s intervals and continue for the duration of the first-column separation. Typically, several 8

hundred second-column separations are obtained and merged by software to generate a two-dimensional chromatogram in which detector data are plotted on a two-dimensional retention plane rather than on a simple time axis. This is represented schematically in Figure 1.5 where the detector sees a continuous stream of one-dimensional data that is then split and rotated based on the modulation period before the software merges the slices to create the final contour plot.

Figure 1.5 Schematic of data processing performed in GC GC showing chromatograms being chopped and merged to display contour plots based on modulation period.

Detectors with fast response times are required due to the sharp bands (200-500 ms) produced by the fast second column separation.19 Although various

detectors have been developed for high speed GC, such as flame ionization detectors and

quadrupole mass spectrometers,24 ToFMS can acquire data for a full mass range at rates fast enough for GC GC while quadrupoles usually need to be run in single ion monitoring mode when coupled with GC GC. ToFMS can track very narrow peaks, allows for automated peak finding, and the spectra are not concentration dependent because the ionization is pulsed.20, 24

Retention Time (s) 2 4 6

n= 6 n= 5 n= 4 n= 3 n= 2 n= 1 n= 0

C4

C10 C8 C6

C13 C12 C11

C14

C17 C15C16

C18

C20 C21

C C22 C23 24 C14 C17 C15C16 C18

C20C21

C C22 C23 24

C4

C10 C8 C6

C13 C12 C11

(A)
0

(B)
0

300

1300

2300

3300 300 Retention Time (s)

1300

2300

3300

Figure 1.6 GC GC chromatogram of fatty acid methyl esters emphasizing the elution of homologous series along an arc (A) and clustered elution of compound classes (B). The clustered elution of C18 and C20 FAMEs is highlighted by the white ovals. n, number of double bonds.

Besides higher resolution, the two-dimensional separation plane of GC GC allows for structured chromatograms in which compound classes have characteristic patterns. Homologous series of analytes tend to elute in characteristic lines (or curves) and compound classes tend to elute in clusters, both of which can be easily recognized. An example of this structure can be found in Figure 1.6 where the fatty acid methyl esters (FAMEs) with the same number of double bonds elute along the same arcs (A) and the FAMEs with the same number of carbons are clustered together (B). The structured nature of the chromatograms assists in classification and identification of components even in the absence of pure standards. 10

Peak Capacity in GC GC Peak capacity in GC GC (ncGC GC) is generally assumed to be equal to the peak capacity of the first dimension column ( nc1) times the peak capacity of the second dimension column (nc2), Equation 1.9. (1.9) Under this assumption, a first dimension column with a peak capacity of 250 coupled in series to a second dimension column with a peak capacity of 10 would provide a twodimensional peak capacity of 2500. In reality though, the actual peak capacity of GC GC is always less than . One reason for this lower peak capacity is because

modulation causes some peak broadening in the reconstructed first dimension. Even under conditions where the peak capacities in both dimensions are optimized, peaks in the first dimension can be 23 % wider with modulation than without effectively lowering nc1.25 Additionally, useful peak capacity in the second dimension is often reduced by the use of columns with film thicknesses and column temperatures that lead to the smallest capacity factors being close to 1.5, a choice that leaves an empty portion of second dimension separation space.25 Because the performance of columns in two-dimensional systems is not directly equivalent to the performance of the stand alone columns, making it difficult to directly compare peak capacities, other metrics have been developed to more directly compare one-dimensional and two-dimensional systems. One of the metrics is the concept of peak capacity gain (Gn ) that results from the addition of the second dimension to a onedimensional system. Gn is described simply in Equation 1.10 25

11

(1.10) where n c1,o is the peak capacity of the stand alone first dimension column. However, because we know that the actual peak capacity of GC GC is always less than nc1 nc2, a more accurate description of peak capacity gain is found in Equation 1.11 25

(1.11)

where N1 is the first dimension column plate number and Rs,min,1 is the lowest acceptable resolution in the first dimension. This equation accounts for losses in both dimensions from an optimal GC GC system and assumes the data analysis of the stand alone first dimension is applied to the second dimension of GC GC. A system that is estimated to be optimal 25 can be considered here to calculate the best possible peak capacity gain in GC GC. In this system the first dimension column is 30 m long with a 250 m i.d. and a 0.1 mm stationary phase film thickness and the second column is 0.86 m long with the same i.d. and film thickness as the first dimension column. When Rs,min,1 is 1.5 Gn,opt is about 9.3 and when Rs,min,1 is 1, Gn,opt is about 14. Unfortunately, current technology does not allow for GC GC systems to run perfectly optimized. For example, the current modulator technology injects pulses on the second column that are widths an order of magnitude higher than optimal.26 When the peak capacity gain from using a current GC GC system instead of the one-dimensional equivalent (defined as a one-dimensional system that has the same analysis time and the same minimum detectable concentration as the GC GC system in question) the Gn,eq

12

(no longer optimal values, therefore now referred to as peak capacity gain equivalent) values drop to about 3 and 4 for Rs,min,1 values of 1.5 and 1, respectively.25 Despite the shortcomings of current GC GC technology to fully utilize the potential peak capacity gain over equivalent one-dimensional separations, the technique still has great analytical importance and is unmatched by conventional GC for many, but not all, complex samples. Additional disadvantages that must be weighed when choosing to use GC GC or GC are costs of commercial instruments (several hundred thousand dollars) and the cryogenics required. Computing power is another necessity because chromatogram files can exceed 2 GB when processed and, if computing power is low, can take hours to process. User experience must also be considered due to the increased complexity and high maintenance requirements of current instrumentation. Dissertation Overview The goal of this research project was to utilize the separation and detection power of GC GC and apply it to novel applications in the areas of geology and metabolomics. All experiments were performed using the commercially available Leco Pegasus 4D which is an Agilent 6890 gas chromatograph modified for comprehensive two-dimensional gas chromatography and coupled to a Pegasus time-of-flight mass spectrometer. Chapter 2 describes both a metabolite profiling and metabolomics analysis of extracts from INS-1 cells incubated in 3 mM, 7 mM and 17 mM glucose. Chapter 3 discusses the use of GC GC-ToFMS to analyze the total lipid content of INS-1 cell extracts incubated in 0 mM, 0.5 mM, 10 mM and 20 mM glucose. Chapter 4 describes

13

the use of pyrolysis-GC GC to analyze Mediterranean Sea sediments, known as sapropels, with high total organic carbon concentrations. Finally, Chapter 5 summarizes and describes future directions for the work completed in Chapters 2 through 4.

14

References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) Shellie, R.; Mondello, L.; Marriott, P.; Dugo, G. J. Chromatogr. A 2002, 970, 225-234. Veriotti, T.; Sacks, R. Anal. Chem. 2001, 73, 4395-4402. Grall, A. J.; Zellers, E. T.; Sacks, R. D. Environ. Sci. Technol. 2001, 35, 163-169. Merola, G.; Gentili, S.; Tagliaro, F.; Macchia, T. Analytical and bioanalytical chemistry, 397. Lowe, R. H.; Barnes, A. J.; Lehrmann, E.; Freed, W. J.; Kleinman, J. E.; Hyde, T. M.; Herman, M. M.; Huestis, M. A. J. Mass Spectrom. 2006, 41, 175-184. Paik, M.-J.; Ahn, Y.-H.; Lee, P. H.; Kang, H.; Park, C. B.; Choi, S.; Lee, G. Clin. Chim. Acta 2010, 411, 1532-1535. Sacks, R. D.; University of Michigan: Ann Arbor, MI, pp 79. Golay, M. J. E. Nature 1958, 182, 1146-1147. Grob, R. L.; Barry, E. F.; Wiley, I. Modern practice of gas chromatography ; Wiley-Interscience: Hoboken, N.J., 2004. Giddings, J. C. Unified separation science ; Wiley: New York, 1991. Mondello, L.; Lewis, A. C.; Bartle, K. D. Multidimensional chromatography ; Wiley: West Sussex, England ; New York, 2002. Adam, F.; Bertoncini, F.; Coupard, V.; Charon, N.; Thiebaut, D.; Espinat, D.; Hennion, M. C. J. Chromatogr. A 2008, 1186, 236-244. Farwell, C.; Reddy, C. M.; Peacock, E.; Nelson, R. K.; Washburn, L.; Valentine, D. L. Environ. Sci. Technol. 2009, 43, 3542-3548. Cordero, C.; Bicchi, C.; Joulain, D.; Rubiolo, P. J. Chromatogr. A 2007, 1150, 37-49. Mohler, R. E.; Tu, B. P.; Dombek, K. M.; Hoggard, J. C.; Young, E. T.; Synovec, R. E. J. Chromatogr. A 2008, 1186, 401-411. Kusano, M.; Fukushima, A.; Kobayashi, M.; Hayashi, N.; Jonsson, P.; Moritz, T.; Ebana, K.; Saito, K. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences 2007, 855, 71-79. Liu, A.; Phillips, J. B. J. Chromatogr. Sci. 1991, 29, 227-231. Wang, Y.; Qian, C.; Norwood, D.; McCaffrey, J. J. Liq. Chromatogr. Rel. Technol. 2010, 33, 1082-1115. McGuigan, M.; Waite, J. H.; Imanaka, H.; Sacks, R. D. J. Chromatogr. A 2006, 1132, 280-288. Dimandja, J. M. D. Anal. Chem. 2004, 76, 167A-174A. Adahchour, M.; Beens, J.; Brinkman, U. A. T. J. Chromatogr. A 2008, 1186, 67108. Cortes, H.; Winniford, B.; Luong, J.; Pursch, M. J. Sep. Sci. 2009, 32, 883-904. Marriott, P.; Shellie, R. Trac-Trends in Analytical Chemistry 2002, 21, 573-583. Song, S. M.; Marriott, P.; Wynne, P. J. Chromatogr. A 2004, 1058, 223-232. Blumberg, L. M.; David, F.; Klee, M. S.; Sandra, P. J. Chromatogr. A 2008, 1188, 2-16. Blumberg, L. M. J. Sep. Sci. 2008, 31, 3352-3357.

(17) (18) (19) (20) (21) (22) (23) (24) (25) (26)

15

Chapter 2 METABOLITE PROFILING AND METABOLOMIC ANALYSIS OF INS-1 CELLS USING COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAPHY TIME-OF-FLIGHT MASS SPECTROMETRY Introduction A thorough understanding of systems biology is important to discover biomarkers and disease mechanisms. Systems biology is comprised of genomics, transcriptomics, proteomics and metabolomics, all of which are complimentary to each other. The metabolome can be defined as the quantitative complement of all the low-molecular weight molecules (<3000 m/z) present in cells in a particular physiological or developmental state. Studying the metabolome has provided the understanding that changes in individual enzyme levels can have significant effects on the concentration of a variety of individual metabolites even if there is little effect on metabolic fluxes. 1 The benefits of metabolomics over transcriptomics and proteomics include the fact that the metabolome is further down the line of gene to function and reflects more closely the activities of the cell at a functional level. Changes in the metabolome are downstream results of gene expression and thus may be amplified relative to changes in the transcriptome and proteome.1 Additionally, unlike transcripts or protein identifications, metabolites are not organism specific and therefore the analytical protocol used to measure a specific metabolite is applicable to fungi, prokaryotes, plants and animals.2 The field of metabolomics has been broken down into four different strategies; metabolite target analysis, metabolite profiling, metabolomic analysis and metabolic 16

fingerprinting. Metabolite target analysis is restricted to metabolites of a particular system that would be directly affected by abiotic or biotic perturbation.2 Metabolite target analysis is usually accomplished with gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), or high performance liquid chromatography (HPLC).2 Metabolite profiling analysis is focused on a group of metabolites, such as those associated with a specific pathway, and metobolomics is the comprehensive analysis of the whole metabolome under a given set of conditions.1 Both metabolite profiling and metabolomics can be completed using comprehensive two-dimensional gas chromatography coupled to mass spectrometry (GC GC-MS), HPLC-MS, LC-MS, or LC coupled to nuclear magnetic resonance (LC-NMR).2 Metabolic fingerprinting is the classification of samples on the basis of either their biological relevance or origin and often involves NMR, direct infusion electrospray ionization MS (DIMS), laser desorption ionization MS (LDI-MS), fourier transform infrared spectroscopy (FT-IR), and Raman spectroscopy.1, 2 In this work we use GC GC time-of-flight MS (ToFMS) to perform both a metabolite profiling and a metabolomics analysis of INS-1 cells. INS-1 cells are a clonal cell line often used as a model for the pancreatic -cell. -cells are one of the four major cell types found in the islets of Langerhans, which are islands of cells found in the pancreas of mammals.3 -cells secrete insulin in response to glucose as well as other nutrients, hormones and nervous stimuli.3 Type 2 diabetes is characterized by the development of early insulin resistance and the fai lure of -cells to compensate with hyperinsulinemia.3 Failure of the -cell is crucial to development of type 2 diabetes.

17

Better understanding of normal and dysfunctional metabolism in these cells may be expected to give insight into -cell failure in type 2 diabetes.

Figure 2.1 Schematic of glucose stimulated insulin secretion (GSIS). Glucose enters the cell through the glucose transporter, triggers glycolysis and mitrochondrial respiration which leads to an increase in the ATP/ADP ratio resulting in closure of the ATP-sensitive K+-channel (KATP). The resultant membrane depolarization opens the voltage-dependant Ca2+-channel and allows a flux of calcium into the cell triggering exocytosis of insulin.4

Glucose stimulated insulin secretion (GSIS) is metabolically driven as outlined in Figure 2.1. Glucose enters the cell through the glucose transporter and triggers glycolysis and mitochondrial respiration which leads to an increase in the ATP/ADP ratio in the cell and results in closure of the ATP-sensitive K+-channel (KATP). The resultant membrane depolarization opens the voltage-dependent Ca2+-channel and allows a flux of calcium into the cell, triggering exocytosis of insulin.5 Despite the overwhelming data supporting this mechanism for GSIS, there is strong evidence that additional KATP

channel-independent pathways exist, evidence of which is demonstrated in Figure 2.2.6 In Figure 2.26 the response of test cells diverges from that of the control cells approximately 6 minutes after exposure to 250 M diazoxide and KCl. Diazoxide prevents KATP channel 18

operation, therefore the continued release of insulin must be independent of the KATP channel. A schematic of the KATP dependent and KATP channel-independent pathways can be found in Figure 2.3.7

Figure 2.2 Evidence of KATP-channel independent pathway. The open squares are control cells and the closed squares are test cells. Both the control and the test cells were equilibrated by exposure to Krebs Ringer HEPES buffer (KRHB) containing 2.8 mM glucose for 40 min. AT the ten minute time point both samples were exposed to 250 M diazoxide and 40 mM KCl at the same time point the test cells were exposed to KRHB containing 16.7 mM glucose. After 6 minutes the insulin secretion of the cells diverged, the control cells slowly decrease and the test cell show an increased rate of release despite the elimination of the KATP-dependant pathway by the diazoxide. (Used with permission from6)

Although glucose is required for normal insulin secretion, excessive glucose can lead to glucotoxicity and -cell dysfunction. Once the primary pathogenesis of diabetes is established, hyperglycemia ensues and exerts additional damaging, toxic effects on the -cell.8 It has been proposed that continuous overstimulation of the -cell by glucose could eventually lead to depletion of insulin stores, worsening of hyperglycemia, and deterioration of -cell function.8, 9 -cell lines can be used as a model for glucotoxicity by exposing the cells to media containing high concentrations of glucose for extended 19

Figure 2.3 KATP channel-dependent (left) and KATP channel-independent (arrows on right) glucose signaling pathways in the -cell are shown. Glucose is transported into the -cell and is metabolized by a cascade of reactions. The metabolic signals give multiple pathways leading to insulin exocytosis. VDCC, L-type voltage-dependent Ca2+ channel; [Ca2+]i, cytosolic free Ca2+ concentration. (Used with permission from7)

periods of time.8,

10

For example, a previous study has shown that cells cultured in

0.8 mM glucose for a prolonged period (multiple passes over several weeks) maintained insulin content and GSIS while identical cells cultured in 11.1 mM glucose had drastically compromised insulin content and GSIS.8 Specifically, INS-1 cells have been used to show that glucotoxic -cells have additional, more distal defects in the exocytotic pathway,8, 10, 11 that glucotoxicity alters calcium handling in cells, and that glucotoxicity alters the expression of several key proteins in exocytosis. 10 Thus, an analysis of the INS-1 metabolome may help identify pathways that are activated during hyperglycemia and glucotoxicity and lead to a better understanding of type 2 diabetes. 11 In this study, we used GC GC to determine metabolite changes that occur as a function of increasing glucose from 3 to 7 to 17 mM. We use both metabolite profiling 20

and undirected metabolomics analysis. The metabolite profiling analysis focuses on the glycolysis and mitochondrial respiration step of GSIS by targeting metabolites amendable to GC in the citric acid cycle and glycolysis as shown in Figure 2.4.

Glycolysis Glucose

Glucose-6-phosphate (G6P)

Ribose-5-Phosphate (R5P)

Fructose-6-phosphate (F6P)

Pentose Phosphate Shunt

Fructose-1,6-bisphosphate (FBP)

Citric Acid Cycle Glyceraldehyde-3-phosphate

+

Dihydroxyacetone phosphate Malate 1,3-Bisphosphoglycerate (1,3-BPG) Fumarate 3-Phosphoglycerate (3PG)

OAA

Citrate

Isocitrate

2-Phosphoglycerate (2PG)

Succinate

Phosphoenolpyruvate Succinyl CoA Pyruvate

a-ketoglutarate

Figure 2.4 Schematics of glycolysis, the pentose phosphate shunt, and the citric acid (TCA) cycle. When glucose enters the cell, glycolysis is initiated and the glucose is metabolized to pyruvate which enters the TCA cycle. 4

Additionally, amino acids can feed into pathways of glucose oxidation and anaplerosis; thus amino acids were profiled as well.12 In this work we double the number of target analytes identified when compared to a previous GC/MS study of INS-1 cells stimulated

21

by glucose and show that our results are in good agreement with what has been observed previously.13 While the metabolite profiling experiments allowed us to examine the reproducibility of the method and ensure that it provided results consistent with known metabolic changes, the metabolomics method allowed us to identify changes in additional metabolites under the experimental conditions. Such changes provide clues to metabolic pathways associated with insulin secretion and allowed for the detection of changes at supramaximal (for insulin secretion) glucose concentrations. As discussed earlier, such changes may help to identify pathways associated with glucotoxicity. GC GC was used for this work because it is capable of both metabolite profiling and metabolomic analysis. Additionally, the increased detectability and increased peak capacity provide distinct advantages when compared to other methods used for metabolite analysis. Metabolomics using GC GC is a rapidly emerging area of study; however, prior to this work, it has not been applied to insulin secreting cells. Previous work includes analysis of metabolites in rye grass samples, 14 urine,15 blood plasma,16, 17 mouse spleen tissue extracts,18, 19 rice,20 and yeast cells. 21, 22 Experimental Reagents All chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise noted. Roswell Park Memorial Institute (RPMI) media, fetal bovine serum (FBS), HEPES, and penicillin-streptomycin were purchased from Invitrogen Corp. (Carlsbad, CA). Cell lifters and 10 cm polystyrene non-pyrogenic culture dishes were 22

purchased from Corning (Lowell, MA). Methoxyamine hydrochloride, pyridine, HPLC grade methanol, citrate, 4 dram screw cap vials, 2 mL autosampler vials and 200 L inserts were purchased from Fisher Scientific (Fairfield, NJ). Ornithine was from Acros Organics (Morris Plain, NJ). D6 -succinate was from Cambridge Isotopes (Andover, MA). Samples INS-1 cells were cultured on 10 cm plates in RPMI-1640 (+l-glutamine) supplemented with 10 % FBS, 1 mM pyruvate, 10 mM HEPES, 50 M 2--mercaptoethanol, and 1 unit penicillin-streptomycin. INS-1 cells were grown to confluence (~4 x 107 cells) in 10 cm polystyrene dishes with RPMI culture media. All cells used in a particular experiment were seeded at the same time, taking care to minimize variability by using precise volumes of reagents and seed cells. Krebs-Ringer-HEPES buffer (KRHB) was prepared to contain 3 mM glucose, 20 mM HEPES, 118 mM NaCl, 5.4 mM KCl, 2.4 mM CaCl, 1.2 mM MgSO 4 , and 1.2 mM KH2 PO4, and adjusted to pH 7.4 with HCl. Cells were washed once with 10 mL of KRHB prior to incubation in 10 mL of KRHB for 30 min. The KRHB glucose concentration was then left at 3 mM or raised to 7 or 17 mM for 18 min at 37 C. Each glucose concentration was prepared in quadruplicate, however, only three plates were available for 3 and 17 mM. After treatment, cells were washed once with 10 mL milli-Q water and snap frozen with liquid nitrogen. Plates were stored at -80 C until extraction. Extraction and Derivatization Extraction was performed by adding 700 L of ice cold 80:20 methanol water to each plate and scraping for approximately 1 min. Samples were then transferred to

23

4 dram glass vials and dried on an Eppendorf Vacufuge (Hauppauge, NY) at room temperature for 1.5 h. After drying, samples were capped and stored at -80 C until derivatization. All samples were derivatized within 24 h of being analyzed by GC GC. Derivatization was performed by warming the extracted samples to room temperature, d6 -succinate and 13 C-glucose were added such that the final concentration of each would be 30 M in the final derivatization volume of 130 L and the samples were placed on the vacufuge for 20 min. Fifty L of 20 mg/mL methoxyamine hydrochloride in pyridine was added and samples were incubated at 30 C for 1.5 h. 80 L of Regisil (BSTFA with 10% TMCS) was then added to each sample followed by incubation at 70 C for 50 min. Samples were allowed to shake at room temperature for 1.5 h and then transferred to 2 mL autosampler vials with 200 L inserts. Comprehensive Spectrometry GCGC analysis was performed on a Leco Pegasus III with 4D upgrade (St. Joseph, MI). The primary column was a 30 m Rxi-1ms (0.25 mm i.d., 0.18 m film) and the secondary column was a 2 m Rtx -200 (0.18 mm i.d., 0.2 m film) both from Restek Corporation (Bellefonte, PA). A 1 L injection was made with an Agilent 7683 automatic liquid sampler (Palo Alto, CA) in splitless mode and five replicates were completed for each sample. The primary oven was maintained at 70 C for 0.5 min and then increased at a rate of 3 C per minute to 250 C and maintained for 5 min. The secondary oven and the thermal modulator were offset from the primary oven by 5 C and 30 C respectively. A modulation period of 7 s was used and the hot pulse time (length of time the hot jet fires to initiate injection on the second dimension) was 0.6 s. A flow rate of 1 mL/min 24 Two-Dimensional Gas Chromatography Time-of-Flight Mass

ultra-high purity helium with an inlet and mass spectral transfer line temperature of 250 C and 300 C, respectively, were maintained. A mass range of m/z 45 to 1000 was collected at a rate of 200 spectra/s after a 390 s solvent delay. The ion source was maintained at 200 C. Preparation and Analysis of Standards A stock solution containing ornithine, ribose-5-phosphate (R5P), glucose-6phosphate (G6P), 3-phosphoglycerate (3PG), pyruvate, lactate, citrate, isocitrate, fumarate, succinate, frucutose-6-phosphate (F6P) and malate was prepared in milli-Q water. Aliquots were transferred to 4 dram vials such that the final concentration would be 30 M (after derivatization). A commercially available amino acid standard (Sigma-Aldrich, St. Louis, MO) was also diluted and transferred to 4 dram vials; the final concentration was again 30 M in 130 L final volume. Standards were evaporated to dryness, derivatized using the same methodology and analyzed under the same chromatographic conditions as the INS-1 cell extracts. Data Analysis Leco ChromaTOF version 4.22 was used for instrument control and data processing. Identification of target analytes was completed through mass spectral library searches and comparison to metabolite standard retention times. The National Institutes of Standards and Technology (NIST) mass spectral library (version 2.0) and a library obtained from the Max Planck Institue of Molecular and Plant Physiology (http://wwwen.mpimp-golm.mpg.de/02-instUeberInstitut/04instRessources/webbasedRsrc/metaboliteMSL/index.html) were used. A similarity

threshold of 700/1000 between a library mass spectrum and an analyte mass spectrum 25

was considered a match. This value was determined to be sufficiently high to minimize the number of false positives while also limiting the number of false negatives. Retention time shifts within 1 modulation period, in the first dimension, were allowed and 0.2 seconds, in the second dimension, as this was on the order of the typical second dimension peak widths.18 Statistical significance was determined in GraphPad Prism version 3.03 (La Jolla, CA) using a one way ANOVA analysis and a Newman-Keuls post-hoc test. All statistical analysis was performed using the ratio of peak area to d 6 -succinate peak area except for glucose which was analyzed using
13

C-glucose instead of d6 -succinate.

Metabolite mapping was performed using the Metscape23 plug-in for Cytoscape.24, 25 Results Metabolite Standards A commercially available amino acid standard containing 30 M alanine, valine, glycine, serine, methoinine, aspartate, proline, threonine, isoleucine, phenylalanine, glutamine, lysine, tyrosine, cystine, arginine and histidine was derivatized and analyzed via GC GC to verify detectability and determine retention times of the available amino acids. As Figure 2.5 illustrates, all analytes were detected and identified except histidine and arginine. Cystine was detected but is not shown in Figure 2.5 for clarity. Glycolysis and citric acid cycle metabolites amendable to GC were also analyzed to verify detectability and determine retention times, and a representative chromatogram can be found in Figure 2.6. 2-Phosphoglycerate (2PG), data not shown, was also detected when a sample containing only 2PG and 3PG was run.

26

Figure 2.5 Total ion chromatogram of the commercially available amino acid standard. Alanine (a), valine (b), leucine (c), proline (d), isoleucine (e), glycine (f), serine (g), theronine (h), methionine (i), aspartate (j), phenylalanine (k), glutamine (l), tyrosine (m), lysine (n).

Figure 2.6 Total ion chromatogram of glycolysis and TCA standards. Lactate (a), succinate (b), fumarate (c), malate (d), 3PG (e), citrate (f), isocitrate (g), ornithine (h), R5P (i), F6P (j), G6P (k).

27

In LC studies of similar metabolites citrate and isocitrate are often reported as one peak, as are G6P and F6P. Using GC GC we are able to separate these isomers. All detectable standard analytes plus glucose were used as target metabolites during analysis of the INS-1 cell extracts. Calibration curves were created for 15 standard analytes. Linear correlation coefficients of 0.99 or greater were achieved in the range of 1 to 30 M. A representative calibration curve for proline is shown in Figure 2.7.

Peak Area (x 106)

5
4 3 2 1 R = 0.9993

0
0 10 20 30 Concentration (M)
Figure 2.7 Calibration curve for proline. Samples were analyzed in triplicate and error bar is standard deviation.

Metabolite Profiling Figure 2.8 illustrates a glucose-stimulated insulin secretion dose-response curve from INS-1 cells. As can be seen here, cells were stimulated at glucose concentrations that correlate with low (3 mM), moderate (7 mM), and high (17 mM) insulin release. Twenty seven of thirty target metabolites (listed in Table 2.1) were detected in INS-1 cell extracts at all three glucose concentrations. Lysine was not detected at 3 mM glucose. It was also only detected in one 7 mM sample and one 17 mM sample. Isocitrate and cystine were not detected in any of the INS-1 cell extracts. A representative total ion 28

chromatogram (TIC) can be found in Figure 2.9 (top) with additional ion channels and zoomed in portions (boxes below TIC) demonstrating the location of the twenty seven targets. The average relative standard deviations (RSDs) for repeat, 1 L, splitless injections of the same extract, or technical RSDs, were 18, 17 and 14 %. The technical RSD of the 30 M amino acid standard and glycolysis/tca standard were 12 % and 13 %, respectively. The average RSDs of each successful injection at a given glucose concentration, or biological RSDs, were 27, 24 and 20 % for 3 mM, 7 mM and 17 mM glucose respectively. The technical and biological RSDs for each analyte, at each glucose concentration, can be found in Tables 2.2 and 2.3. Metabolite levels at each glucose concentration were compared using one-way ANOVA.

Insulin Secreted (% maximum)

100 80 60 40 20 0 0 5 10 15 20

Glucose (mM)
Figure 2.8 Glucose dose-response curve where maximal insulin release occurs at ~10 mM glucose. Data was obtained from 10cm plates of INS-1 grown to ~ 70% confluence (31 MM cells) in RPMI. Media changed to low glucose RPMI (3mM) for ~20 hr prior to experiment. Media changed to KRB (no glucose + 0.2% FAF BSA) and spiked to indicated glucose concentration. Media removed for insulin measurement and metabolism quenched at30 min. Error bars are SEM. n=3. Data and figure courtesy of Matthew Lorenz.

29

Table 2.1 List of target metabolites.

Ornithine Ribose-5-Phosphate Glucose-6-Phosphate 3-Phosphoglycerate 2-Phosphoglycerate Pyruvate Lactate Citrate Isocitrate Fumarate Succinate Fructose-6-Phosphate Malate Cystine

Alanine Valine Glycine Serine Methionine Aspartate Proline Threonine Isoleucine Phenylalanine Glutamine Lysine Tyrosine

30

350

1350

2350

3350

4.4

H G

4.0

L
M

Retention Time (s)

B
A

I E

2.4

3.0

490
3.0

990
3.4

1490

1477

1677
5.1

1877

P
A

2.9

2.6

490
3.6

640
3.6

2263

2363
3.8

4.8

2260

2290

U
V

2.6

3.2

2459

2559

2495
4.5

2525
6.5

2.4

2431

2700

4.7

X X Y

4.3

4.1

2867

2907

3196

3236

6.1

2185

2215

Retention Time (s)

Figure 2.9 Total ion chromatogram of a 17 mM INS-1 cell extract (top). Analytes of interest are highlighted for clarity. Lactate (A), valine (B), leucine (C), proline (D), isoleucine (E), glycine (F), succinate (G), fumarate (H), serine (I), threonine (J), malate (K), methionine (L), aspirate (M), glutamate (N), phenylalanine (O), pyruvate (P), citrate (Q), ornithine (R), 3PG (S), lysine (T), tyrosine (U), glucose (V), R5P (W), F6P (X), G6P (Y), 2PG (Z).

31

Table 2.2 Technical relative standard deviations for metabolite profiling analysis

3 mM Glucose 7 mM Glucose 17 mM Glucose Alanine 11 18 16 Valine 12 11 6.5 Proline 13 18 13 Glycine 16 17 16 Fumarate 11 9.1 11 Threonine 13 16 12 Malate 6.4 10 7.1 Methionine 31 23 20 Aspartate 20 25 16 Phenylalanine 13 27 25 Ornithine 50 42 22 Citrate 11 9.8 20 Tyrosine 18 23 22 R5P 33 8.5 18 F6P 29 13 14 G6P 25 25 12 Succinate 7.4 9.6 4.1 Glutamine 24 11 12 Isoleucine 12 16 14 Leucine 7.8 24 9.9 Lactate 7.45 18 17 Pyruvate 34 12 9.5 Serine 36 25 9.7 Glucose 2.8 9.5 18 3PG 5.2 9.8 10 2PG 12 11 10

32

Table 2.3 Biological relative standard deviations for metabolite profiling analysis.

3 mM Glucose 7 mM Glucose 17 mM Glucose Alanine 12 20 1.9 Valine 4.6 26 18 Proline 13 16 4.9 Glycine 5.4 20 14 Fumarate 13 14 12 Threonine 7.2 36 19 Malate 35 27 14 Methionine 37 24 39 Aspartate 37 7.4 22 Phenylalanine 72 39 38 Ornithine 64 36 46 Citrate 19 26 20 Tyrosine 39 63 43 R5P 37 14 32 F6P 58 22 20 G6P 43 31 24 Succinate 13 12 11 Glutamine 44 13 8.0 Isoleucine 25 20 32 Leucine 17 43 16 Lactate 15 13 5.1 Pyruvate 5.5 27 19 Serine 49 56 18 Glucose 3.2 3.6 25 3PG 13 4.4 8.0 2PG 7.2 11 10

33

Metabolomic Analysis Fisher Ratio anlaysis was used to perform a metabolomic analysis of all analytes changing between each glucose concentration, that is, from 3 mM glucose to 7 mM glucose, from 3 mM glucose to 17 mM glucose and from 7 mM glucose to 17 mM glucose. A Fisher Ratio is defined as the class-to-class variation of the detector signal divided by the sum of the within-class variations of the detector signal and is calculated using Equation 3.115
(3.1)

where cl is the class-to-class variation and err is the within-class variation. cl and err are described in Equations 3.2 and 3.3 15

(3.2)

(3.3) Where ni is the number of measurements in the ith class, is the overall mean, h is the number of classes, and N is the total number of sample profiles.15 One advantage of using the Fisher Ratio calculation is that the calculation is robust against biological diversity because it differentiates class-to-class variation from within-class variation. Additionally, unlike other statistical methods, it does not just consider a subset, such as the TIC or a single mass channel, of the 4D data generated by is the mean of the ith class,

is the ith measurement of the jth class

34

GC GC-ToFMS. The Fisher Ratio calculation considers all of the data simultaneously and objectively identifies the most significant differences between complex samples. 21 A significant Fisher Ratio must be determined by the analyst;26 therefore, to establish a reasonable threshold for this data set, all of the 3 mM glucose data was randomized into two groups and a Fisher Ratio analysis performed the results of which can be found in Figure 2.10(A).

Figure 2.10 Fisher Ratio plots for randomized 3 mM glucose groups (A), 3 mM glucose compared to 7 mM glucose (B), 3 mM glucose compared to 17 mM glucose (C) and 7 mM glucose compared to 17 mM glucose (D). The red line in (A) indicates the 1064 threshold.

Aside from a few artifacts, there is little change between analytes when the two randomized 3 mM groups are compared. A value of 1064 was chosen as the threshold for this work, i.e. any compounds with a Fisher Ratio of greater than 1064 were considered to be changing significantly between the glucose concentrations. This threshold is shown

35

on the 3 mM versus 3 mM glucose and 3 mM versus 17 mM glucose Fisher Ratio plots in Figure 2.11. As shown on the histograms in Figure 2.12, the threshold of 1064 excludes all but 4 % of the data from the 3 mM versus 3 mM glucose analysis and 34 % from the 3 mM versus 17 mM data, thus providing a minimum number of both false positives and false negatives.

Figure 2.11 Zoomed in Fisher Ratio plots for 3 mM versus 3 mM glucose (left) and 3 mM versus 17 mM glucose (right) with red line indicating the 1064 threshold.

Of the 3128 peaks detected in the 3 mM versus 3 mM glucose data set, 135 (4 %) are above the 1064 threshold which is highlighted in Figures 2.11 and 2.12. However, only 38 of the peaks were identified by the software and 8 of those peaks can be attributed to column bleed or the derivatization agent. Therefore, approximately 1 % of the peaks detected in the 3 mM versus 3 mM data were significantly changing and are potential false negatives at the 1064 threshold. Contributions from column bleed and derivatization reagents were also disregarded. Fisher Ratio plots for the analysis of 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose can be found in Figures 2.10(B) through (D). Unlike in 2.10(A), there are 1133 analytes with a Fisher Ratio greater than 1064 in the 3 mM to 17 mM analysis and 882 and 1051

36

in the 3 mM glucose to 7 mM glucose and 7 mM to 17 mM glucose analyses respectively.

4%

34 %

Figure 2.12 Histograms of Fisher Ratios for 3 mM vs 3 mM glucose (top left) and 3 mM vs 17 mM glucose (top right). Rescaled histograms of Fisher Ratios for 3 mM vs 3 mM glucose (bottom left) and 3 mM vs 17 mM glucose (bottom right). Red lines indicate the location of the 1064 Fisher Ratio threshold with only 4 % of the compounds in the 3 mM vs 3 mM glucose data falling above this threshold.

Nine hundred fifty (3 mM to 17 mM), 696 (3 mM to 17 mM) and 873 (7 mM to 17 mM) of the analytes with a Fisher Ratio above 1064 were not identified, i.e. a match between the mass spectrum for the peak and a library mass spectrum of 70 % or greater did not exist. Additionally, if a duplicate was found, the peak with the lower Fisher Ratio was disregarded as well. Kyoto Encyclopedia of Genes and Genomes (KEGG)27-29 identifications were assigned to all analytes for which a KEGG ID existed and analytes with multiple derivatives were only included once. This process, which is summarized in Figure 2.11, left 73, 80, and 65 analytes of interest in the 3 mM to 7 mM, 3 mM to

37

17 mM and 7 mM to 17 mM data sets, respectively. The final lists for each data set, including KEGG ID, direction of change, average areas and area differences, can be found in Appendices A through C.

38

Figure 2.13 Flow chart summarizing the process used for determining analytes of interests in the 7 mM to 17 mM glucose data set after Fisher Ratio analysis. This method was also used for the 3 mM to 7 mM and 3 mM to 17 mM glucose data sets.

39

To interpret the changes in cellular content of these analytes and to identify metabolic pathways that were affected by glucose, we used the Metscape23 plug-in for Cytoscape.24, 25 Cytoscape is an open source software platform for the visualization and analysis of complex data sets such as the metabolomics data acquired in this work. The maps of the metabolic pathways obtained for each of the three data sets are shown in Figures 2.14 through 2.16 where the red dots represent the analytes input by the user (or seeds) and the blue spots represent other metabolites involved in the pathways, the black dots are reactions that the analytes are involved in and the lines connect the related metabolites and reactions to create the map of pathways. The maps each consist of one big network where all of the target analytes are connected, a few small sub networks, and some isolated analytes that are not connected to the pathways involved or for which a pathway does not exist. Each of these components is highlighted in Figure 2.14 for clarity. The analytes were involved in 32 metabolic pathways that can be found in Table 2.4. Additionally, a list of isolated analytes can be found in Table 2.5.

40

n1acetylspermidine agmatine sadenosylmethioninamine (s)1pyrroline5carboxylate proline 5methylthioadenosine RE1537 acetamidopropanal R01920 R01157 R02422 ()ureidoglycolate R01251 R01252 allantoate R01248

lprolyltrna(pro)

trna(pro) R03661

spermidine

trans4hydroxyproline

urea R00670 nacetylputrescine R01154 putrescine lornithine peptide R00135

(B)

R00551

acetoacetate

arginine R01151

succinate

R01364 R00408 R00412 R02164 4fumarylacetoacetate lthreonyltrna(thr)

trna(thr)

R01086 guanidinoacetate R03663

R00565 3methylcrotonylcoa 4aminobutanal 4hydroxybutanoate 10,11dihydro12Rhydroxyleukotriene D4 R01083 threonine fumarate

3methylbutanoylcoa

n(larginino)succinate R01644 lcitrulline

3alphahydroxy5betacholanate 3methylcrotonoylglycine isovalerylglycine RE2111 RE2427 R00751

n6(1,2dicarboxyethyl)amp succinate semialdehyde nacyloacetylneuraminate acetyl phosphate R01954 nacetyllaspartate R01082 lipoylprotein sarcosine n4(acetylbetadglucosaminyl)asparagine R02549 R01986 trna(asp) inosine 5phosphate RE1821 10,11dihydro12Rhydroxyleukotriene E4 RE2638

arachidonylcoa

choloylcoa

glycolithocholate Narachidonoylglycine glycocholate

4hydroxyphenyl acetate

R01648

nformyllaspartate R01135 nacetylbetadglucosaminylamine R03421 saminomethyldihydrolipoylprotein R00367 R00610 R00611 R00488

RE2117

hydroquinone R01810

R00317 laspartyltrna(asp)

R05577 R03425

R03718 rsalanylglycine

rsalanine 2methylpropanoylcoa

nacetylneuraminate 1alkylsnglycerol3phosphocholine 1alkyl2acetylsnglycero3phosphocholine R06893 R00526

acylcoa R04452 formate R01989 acetaldehyde R00710 R01987 acetate 4acetamidobutanoate R00395 R00316 R00489 acetyl adenylate R03647 coenzyme a R00357 R00352 R00351 aspartate R00216 (s)2acetolactate adenosine 5triphosphate glycine R00342 nacylglycine R00214 4aminobutanoate betaalanine laspartyltrna(asn) (s)malate carbon dioxide 4guanidinobutanoate l2amino3oxobutanoate

R04951 RE2429 2methylbutanoylcoa 2amino3oxoadipate R00831 RE2428 glycyltrna(gly) isobutyrylglycine trna(gly) R03654 2methylbutyrylglycine

R00371

R00711

lallothreonine R06171 RE3632 S[2carboxy1(1Himidazol4yl)ethyl]glutathione

R00362 R00235 myoinositol 4phosphate 1dmyoinositol 3phosphate 1phosphatidyldmyoinositol R01900 R01325 cisaconitate 2amino4hydroxy6(derythro1,2,3trihydroxypropyl)7,8dihydropteridine R01186 acetylcoa R01187 isocitrate R01324 citrate R00227

oxaloacetate adenosine 5diphosphate

S[2carboxy1(1 Himidazol4yl)ethyl]Lcysteine R00830 succinylcoa R03284 R00226 R00344 R00200 5aminolevulinate 4trimethylammoniobutanal

fatty aldehyde alkylglycerol snglycero3phospho1inositol

RE3273

phosphatidate inositol 1phosphate

R00210 R00209 R00261 R01185

phosphoenolpyruvate

(s)lactate (r)lactate

3hydroxyn6,n6,n6trimethylllysine RE0596

RE3507

R00196

Prostaglandin PGE2 3glyceryl ester tetrahydropteridine R03331 myoinositol prostaglandin e2 RE2068 R01184 Prostaglandin PGE2 1glyceryl ester RE2067 dglucuronate alphaketoglutarate thiamin diphosphate triacylglycerol RE2265 dglyceraldehyde R01036 R01350 1alphadgalactosylmyoinositol R01041 RE2563 2arachidonoylglycerol glycerol3phosphate RE2754 glycerol R00369 R00847 R01194 stachyose asparagine lalanine 2methyl3oxopropanoate noladin ether R03616 galactosylglycerol R01351 2acylglycerol alphadgalactose 1phosphate raffinose R01095 acylglycerol galactitol R01093 dmannose R01103 R01329 galactan epimelibiose R01105 trna(gln) R03652 lactose glutamine trna(ala) RE1919 salsolinol 1carboxylate dgalactose R00802 betadfructose R02743 R01092 R06098 R01678 NacetylLalanine R00782 R03038 lalanyltrna gamaLglutamylLalanine RE1473 5oxoproline R01104 alphadglucose R00575 RE2031 R03634 R00256 R00253 RE2642 R00894 carbamoyl phosphate R00578 2(alphahydroxyethyl)thiamine diphosphate glutamic acid R00258 R00014 Somatostatin fragment 314 pyruvate somatostatin R00355

R00703 R00704

leukotriene e4

leukotriene d4

4methylthio2oxobutanoate 2acetolactate ncarbamoyllaspartate R01397 sacetyldihydrolipoamide R01699 R00006 R00652

lmethionine

tetrahydrofolate glyoxylate R00497 R00945 rsh thiocyanate R00220 h2o2 R03528 rsch2ch(nh3+)coo3h6no2rs 5,10methylenetetrahydrofolate R01221 R00366

glutathione

R02050

R03105

RE0159

R05861

R01352

gammalglutamyllcysteine (r)3amino2methylpropanoate cyanide serine dalanine

nh3 mercaptopyruvate

R00891

R00896

2oxobutanoate R01001

RE2717

sucrose

glutaminyltrna R01231

R01072

cysgly

R02197 ribulose5phosphate dribose

sphingosine

R05549

R01100

xanthosine 5phosphate

psychosine

guanosine 5phosphate R00257 nad

5phosphoribosylamine

5phosphoalphadribose 1diphosphate lcysteine R00899 RE2223 R00893 3sulfinolalanine R01049

R02926

lcystathionine R01056 3methyl2oxovaleric acid

R01051

R01101

R00768

dgal alpha 1>6dgal alpha 1>6dglucose

melibiose R00801 dglucosamine 6phosphate

dribose 1phosphate ribose 5phosphate R01057 R01054 adpribose

melibiitol

dfructose 1phosphate

deaminonad+ 2oxoglutaramate

4(2aminoethyl)1,2benzenediol NacetylLcysteine dxylulose 5phosphate R02198 R01641

dsorbitol dglucose

R00866 dfructose 6phosphate

R01375

dglyceraldehyde 3phosphate

R00875 RE1342

dfructose

R00892 R00867 betadfructose 6phosphate R00876 RE2811

lcysteinyltrna(cys) R02199 R06590 iminoerythrose 4phosphate dsedoheptulose 7phosphate

R03650 RE2783 RE2781 R00010 R00725 R05804 fructose 3phosphate R01139 R01140 R04230 R02383 R04233 lcystine isoleucine aminofructose 6phosphate

adenosine 5monophosphate phenylpyruvate trna(cys) RE0936 (r)4phosphopantothenoyllcysteine RE1915 5Scysteinyldopa R00689 R00692 R00694 4hydroxyphenylacetaldehyde R02382 tyramine RE2124 nmethyltyramine trna(ile) RE0937 lisoleucyltrna(ile) neocasomorphin (15) betacasomorphin d4phosphopantothenate R03656 betacasomorphin (16)

2deoxyadenosine 5diphosphate dglucose 6phosphate trehalose 2deoxyadenosine 5triphosphate

lphenylalanine

neocasomorphin

R00699 R00698 phenethylamine dopaquinone

lphenylalanyltrna(phe) 2phenylacetamide tetrahydrobiopterin4acarbinolamine R03660 R01795 RE2650 RE0026 R02078 RE2269 trna(phe) RE0938 apelin (112) kinetensin leucine dihydrobiopterin 5,6,7,8tetrahydrobiopterin RE3062 ltyrosine 3,4dihydroxylphenylalanine RE1465 R00736

kinetensin 17 kinetensin 47 RE2268 kinetensin 13

(A)
R02332 R01549 R00968 benzoyl phosphate palmitoylcoa R01274 R00967 R01705 R00970 uridine5monophosphate R01880 R01421 RN0034 palmitate R00963 uridine R02327 R02097 benzoate R01876 R00964 R01706 uracil R05590 R01878 hexadecanoyl[acp] benzamide R02201 delta1piperideine2carboxylate R02203 (9z,11e)(13s)13hydroperoxyoctadeca9,11dienoate 1acylsnglycero3phosphocholine 12(13)epome tetradecanoyl[acp] pipecolate phosphatidylcholine R03626 R02205 palmitoylprotein R02204 9(10)epome R07055 linoleate R07063 R07064 R07056 gammalinolenic acid RE1576 R07057 RE1514 RE1575 11HpODE dodecanoyl[acylcarrier protein] RE3409 acylcarrier protein myristic acid protein 2,3,4,5tetrahydropyridine2carboxylate 9(s)hpode linoleoylcoa lauric acid dxylose R01431 xylitol

apelin13

R00731 R00734 R00729 R03539 R04730

R01815

RE0941 tublin R02918 ltyrosyltrna(tyr)

3(4hydroxyphenyl)pyruvate

3iodoltyrosine

ltyrosyl[alphatublin] iodide trna(tyr)

propanoylcoa 3methyl2oxobutanoate R01214 valine

RE2649

propanoate

R01354

propinol adenylate dgluconic acid

RE0124 6phosphodgluconate butanoate

R01176

phenylethanolamine entkaurene larabinofuranose caprylic acid diethanolamine sebacic acid butanoylcoa hydroxylamine derythrose

pentadecane

azulene

lnorvaline 2trans,6transfarnesol lrhamnose 5oxodproline elaidic acid

dribonate

glutarate

dgalactonate myristoleic acid orthophosphate 2dehydro3deoxydxylonate mannitol

lvalyltrna(val) R01904 R03665

lxylulose R01903 larabitol

trna(val)

(B)

(C)

cytidine

Figure 2.14 Pathway map for 3 mM to 7 mM glucose data set obtained from Metscape.23 The large main network is shown in (A), sub networks are highlighted in (B) and isolated analytes in (C).

40

protein hexadecanoyl[acp] R01761

lribulose

arabinose R01759 R01706

R01705 palmitoylprotein

R01758

larabitol

palmitate R01903

lxylulose 5,6eet 11,12eet R01274 RN0034

14,15eet R01904 R07052

R07050

RE3287

RE3286 RE3289 R07048 8,9eet xylitol

palmitoylcoa 5hydroperoxy6trans8,11,14ciseicosatetraenoate prostaglandin g2 20hete 15hpete 16(r)hete R07041 9peroxy5Z,7E,11Z,14Zeicosatetraenoate R07054 RE3449 n1acetylspermidine nacyloacetylneuraminate 8peroxy5Z,9E,11Z,14Zeicosatetraenoate RE3455 acetyl phosphate 1alkyl2acetylsnglycero3phosphocholine R01281 trna(ser) 9(s)hete anandamide R01810 nacetylneuraminate R01920 5methylthioadenosine 1alkylsnglycerol3phosphocholine R06893 agmatine R01157 hydroquinone R01986 R02549 trna(asn) R01154 acetate 4acetamidobutanoate nacetylputrescine R00316 NacetylLasparagine R01987 R00227 R00235 R00670 acetyl adenylate citrate 4aminobutanoate R01648 n2acetyllornithine R00669 R00362 acetylcoa lglutamate 5semialdehyde R00667 lornithine asparagine R01398 trna(gln) arachidonylcoa R03539 xanthosine 5phosphate RE0941 R02382 R02383 deaminonad+ hmgcoa R01360 ncarbamoyllaspartate ()ureidoglycolate tetrahydrobiopterin4acarbinolamine oxaloacetate R02422 urea R00551 R01397 R00261 5phosphoribosylamine R01954 R00357 formate nformyllaspartate n(larginino)succinate R00526 acetoacetylcoa R03647 R00342 aspartate R00489 betaalanine acetoacetate R01086 laspartyltrna(asp) RE3301 R03660 trna(phe) R01361 3hydroxybutanoic acid R01364 R01135 (s)malate R01083 4fumarylacetoacetate fumarate n6(1,2dicarboxyethyl)amp trna(asp) R01082 R03421 R00488 guanidinoacetate (r)3amino2methylpropanoate R02050 nacetylbetadglucosaminylamine R00214 n4(acetylbetadglucosaminyl)asparagine nacetyllaspartate R00768 dglucosamine 6phosphate RE2642 RE2265 somatostatin phosphatidate leucine 2methyl3oxopropanoate lalanine kinetensin 48 Somatostatin fragment 314 kinetensin 4methyl2oxopentanoate RE2304 RE1473 cysgly R00565 R05577 R00248 R00243 R00216 R02048 R00258 trna(ala) R03038 kinetensin 17 RE2269 kinetensin 13 R00369 gamaLglutamylLalanine kinetensin 47 RE2268 pyruvate lalanyltrna R00355 glutamic acid R00253 R00256 glutamine R01375 2oxoglutaramate tetrahydrofolate R00575 R00578 laspartyltrna(asn) R00220 5phosphoalphadribose 1diphosphate R00945 phenylpyruvate R00689 lphenylalanine glutaminyltrna R03652 R01231 guanosine 5phosphate R00257 nad R01072 5,10methylenetetrahydrofolate RE2117 Narachidonoylglycine R00692 R00698 RE0938 apelin13 R00694 R00699 phenethylamine 2phenylacetamide apelin (112) RE1465 dihydrobiopterin 3,4dihydroxylphenylalanine R01795 RE3062 phosphatidylserine RE2650 R01815 5,6,7,8tetrahydrobiopterin R00731 R02078 RE2124 carbamoyl phosphate ltyrosine 4(2aminoethyl)1,2benzenediol R00736 tyramine RE2781 succinate semialdehyde R01989 R03648 lasparaginyltrna(asn) RE2032 RE2644 R00891 R00585 R00590 2aminoacrylate phosphatidylethanolamine 3(4hydroxyphenyl)pyruvate ltyrosyl[alphatublin] lcysteine 3hydroxypyruvic acid phosphatidylcholine RE0324 trna(tyr) ethanolamine R00582 lcystathionine R01290 R01289 cystathionine R01317 R01151 4aminobutanal lseryltrna(ser) R03662 R04452 R00711 R00710 homocysteine R00317 ophospholserine 7HETE RE3646 8hydroxyeicosatetraenoate acetaldehyde RE3648 RE3650 RE3653 RE3654 RE3651 RE3647 11peroxy5Z,8Z,12E,14Zeicosatetraenoate RE3649 11hydroxyeicosatetraenoate 18HETE 17HETE 13HETE 10HETE arachidonate 15(R)Hydroxy(5Z,8Z,11Z,13E)eicosatetraenoate RE3042 R07046 R01595 R01590 R01593 R01596 19(s)hete 12hpete RE3288 R01431

R07051 dxylose

RE3452 RE1537 3dehydrosphinganine 12peroxy5Z,8Z,10E,14Zeicosatetraenoate RE3458

sadenosylmethioninamine spermidine

acetamidopropanal

4hydroxyphenyl acetate

RE1978

putrescine

4guanidinobutanoate serine RE3298 RE3299 choline

ltyrosyltrna(tyr) tublin R02918 R00734 R00729 R04730 3iodoltyrosine iodide

4hydroxyphenylacetaldehyde

lcitrulline

allantoate

RE0026

dopaquinone nmethyltyramine

R01357

arginine

lphenylalanyltrna(phe)

inosine 5phosphate R02164 R00410 R00408 R00412

4oxoglutaramate R00270 succinate alphaketoglutarate

R01090

RE2270

kinetensin 18 lleucyltrna

peptide 3hydroxylaspartate R04073

RE2031

glycyltrna(gly) trna(gly) NacetylLalanine R03654 R03657 trna(leu)

carbon dioxide saminomethyldihydrolipoylprotein RE2273

peptide laspartate

R03534

R00268 R03425 R00709 R00621 oxalosuccinate sarcosine glycine 3carboxy1hydroxypropylthpp RE0361 isocitrate thiamin diphosphate succinylcoa lipoylprotein dfructose 6phosphate RE3273 R00610 RE2427 R00830 1phosphatidyldmyoinositol 10,11dihydro12Rhydroxyleukotriene E4 3hydroxyn6,n6,n6trimethylllysine 3methylbutanoylcoa R00611 R03718 isovalerylglycine choloylcoa lallothreonine R00367 R06171 glycocholate neuromedin N (14) neuromedin N

2hydroxyglutarate

5aminolevulinate coenzyme a

RE1821 R00371 R03284 10,11dihydro12Rhydroxyleukotriene D4 leukotriene e4 4trimethylammoniobutanal

l2amino3oxobutanoate R00831 R00876 2amino3oxoadipate RE0596 R01140 dglucuronate R00395 RE3632 S[2carboxy1(1 Himidazol4yl)ethyl]Lcysteine rsalanylglycine

R00751 R04951 leukotriene d4

R01184

nacylglycine

RE2428 myoinositol 2deoxyadenosine 5triphosphate 2deoxyadenosine 5diphosphate dfructose 1phosphate RE2638 R01186 R00866 R01185 R01187 fructose 3phosphate RE2783 R01330 R00867 dfructose myoinositol 4phosphate gammalglutamyllcysteine 4methylthio2oxobutanoate RE2111 3methylcrotonoylglycine glycolithocholate glutathione R01221 RE2429 S[2carboxy1(1Himidazol4yl)ethyl]glutathione rsalanine R00497 isobutyrylglycine threonine

acylcoa

2methylbutyrylglycine

2methylbutanoylcoa 2methylpropanoylcoa

R03663

lthreonyltrna(thr) trna(thr)

inositol 1phosphate

1dmyoinositol 3phosphate

3alphahydroxy5betacholanate 3methylcrotonylcoa

nh3 R00366

R00652

betadfructose 6phosphate mannose6phosphate R03331 snglycero3phospho1inositol alkylglycerol dmannose R01194 lmethionine

fatty aldehyde RE3507 inosine 5diphosphate

R01327

R01326 1alphadgalactosylmyoinositol R00875 glyoxylate

2amino4hydroxy6(derythro1,2,3trihydroxypropyl)7,8dihydropteridine tetrahydropteridine R00801

RE2563 2arachidonoylglycerol triacylglycerol R01350 glycerol

R03616

inosine 5triphosphate

R01329 alphadgalactose 1phosphate

R00717 R00465 dsorbitol R00475

R01092 R01352 R00847 galactosylglycerol 2acylglycerol R01351 R01036 R01041 R01104 epimelibiose dgalactose glycerol3phosphate dglucose R02926 sucrose glycolate

acylglycerol

dglyceraldehyde

R01100 RE2067 RE2754 R01095 RE2068 R01093 R01678 sphingosine galactan RE2717 R05549 R03634 R01101 R01105

R01103

melibiitol

R06098 R00802 betadfructose lactose

Prostaglandin PGE2 1glyceryl ester prostaglandin e2

galactitol

raffinose melibiose

alphadglucose

Prostaglandin PGE2 3glyceryl ester

noladin ether

dgal alpha 1>6dgal alpha 1>6dglucose psychosine

stachyose

lprolyltrna(pro) R01248 trna(pro) (s)1pyrroline5carboxylate proline R03661 R00970 R01549

R02332

R00968

benzoyl phosphate

2,3,4,5tetrahydropyridine2carboxylate

4oxo2nonenal

lauric acid

acrylamide valine R05551

maltose 3methyl2oxobutanoate R01214

glucose

propanoylcoa

RE2649

propanoate

R01354

propinol adenylate stearoylcoa

RE0344

stearate

dgluconic acid

dgalacturonate hexanoate RE0124 6phosphodgluconate

lnorleucine 6carboxyhexanoate dfructose 2phosphate 5cholestene entkaurene suberic acid

pentadecane

decanal

tridecane

lnorvaline

xylose

mannitol 2dehydro3deoxydxylonate elaidic acid 2trans,6transfarnesol 5oxodproline lrhamnose

butanal

dribonate

glutarate

dgalactonate

llyxose

oxalate

derythrose

malonate

azulene

1octanol larabinofuranose orthophosphate

R00967 RE3420 uridine5monophosphate R01880 R01421 R02205 R02204 azelaic acid 1hydroperoxy8carboxyoctyl 3,4epoxynon(2E)enyl ether R02327 pipecolate RE3417 tetradecanoyl[acp] uracil R05590 R01878 R02201 3,4epoxynonanal delta1piperideine2carboxylate RE1576 methanethiol R02203 RE1575 dodecanoyl[acylcarrier protein] acylcarrier protein myristic acid 3(methylthio)propionic acid RE3515

R05196

propenoate R03665

lvalyltrna(val) (1,4alphadglucosyl)n

R01251 R01252

R00963

uridine

R02097

benzoate R01876 R00964

trna(val)

trans4hydroxyproline

benzamide

peptide R00135

cytidine

Figure 2.15 Pathway map for 3 mM glucose to 17 mM glucose obtained from Metscape.23

41

2amino4hydroxy6(derythro1,2,3trihydroxypropyl)7,8dihydropteridine fatty aldehyde

lribulose R01549

R02332

R00968

2,3,4,5tetrahydropyridine2carboxylate

R01761 snglycero3phospho1inositol 2arachidonoylglycerol 2acylglycerol

R00967 R02204

arabinose

R01759

R00970

uridine5monophosphate

R01880

R02205

R00963 R01758 tetrahydropteridine RE3507 alkylglycerol inositol 1phosphate RE2563 Prostaglandin PGE2 1glyceryl ester R01352 R03331 dglyceraldehyde R01903 acylglycerol larabitol

uridine R02327 pipecolate R02097 R01876 R00964 R02203

uracil R02201 R01878 delta1piperideine2carboxylate

prostaglandin e2 trna(met) Prostaglandin PGE2 3glyceryl ester RE2067

R01036 R01041

R01351

noladin ether lxylulose

RE2754 sadenosyllmethionine glycerol3phosphate cytidine R01904

RE2068 nacetylmethionine R00847 R01350 lmethionyltrna trna(thr) R03659 glycerol triacylglycerol xylitol

R00177 RE2030 RE2640 R03616

R03663

R00648

lthreonyltrna(thr) lmethionine S[2carboxy1(1Himidazol4yl)ethyl]glutathione 10,11dihydro12Rhydroxyleukotriene D4 2methylbutanoylcoa 3methylbutanoylcoa betaine

galactosylglycerol

4methylthio2oxobutanoate threonine dgal alpha 1>6dgal alpha 1>6dglucose 3methylcrotonylcoa 2methylpropanoylcoa R01104 sphingosine galactan R00946 n,ndimethylglycine R00717 RE2429 R00465 3methylcrotonoylglycine leukotriene e4 R04951 R01678 3alphahydroxy5betacholanate lallothreonineRE0596 rsalanylglycine 2amino3oxoadipate R00611 l2amino3oxobutanoate R00831 RE2638 R06171 R00371 alphadglucose R01100 choloylcoa 4hydroperoxy2nonenal 11HpODE gammalinolenic acid RE3411 nonanoate RE3409 (9z,11e)(13s)13hydroperoxyoctadeca9,11dienoate trna(gly) R03626 12(13)epome R07056 R03654 linoleate RE1514 R07055 phosphatidylcholine R07064 9(10)epome 1acylsnglycero3phosphocholine tetrahydrofolate 5,10methylenetetrahydrofolate 9(s)hpode Narachidonoylglycine R00945 glycine lcystathionine R00801 linoleoylcoa R07057 R01221 acylcoa glycyltrna(gly) dsorbitol 3hydroxyn6,n6,n6trimethylllysine glycocholate R01290 sucrose R03718 R01289 glycolithocholate R03284 cystathionine R00802 R01092 R01101 melibiitol R01103 R00610 homocysteine R00367 glyoxylate lactose R06098 dgalactose R03634 melibiose myoinositol sarcosine R00475 R00652 glycolate R05549 R01105 R01095 R01194 R01093 RE2717 galactitol 1alphadgalactosylmyoinositol psychosine

RE2427 rsalanine R00751 isovalerylglycine

10,11dihydro12Rhydroxyleukotriene E4 S[2carboxy1(1 Himidazol4yl)ethyl]Lcysteine RE3632 RE1821 RE2428 2methylbutyrylglycine

R02821

isobutyrylglycine RE2111

leukotriene d4

stachyose

4trimethylammoniobutanal

raffinose

alphadgalactose 1phosphate betadfructose R02926

R07063

R00366

dglucose

epimelibiose R01329

RE2117 phosphatidate arachidonylcoa RE3299 RE3301 phosphatidylserine serine RE0324 R01317 3dehydrosphinganine 3hydroxypyruvic acid R00582 somatostatin ophospholserine palmitoylcoa succinylcoa R03662 10HETE RE3651 anandamide RE3286 prostaglandin g2 R01590 arachidonate RE3654 17HETE trna(ser) 2(alphahydroxyethyl)thiamine diphosphate 18HETE RE3650 RE3648 RE3449 RE3455 RE3646 9(s)hete R00200 R07051 8hydroxyeicosatetraenoate 9peroxy5Z,7E,11Z,14Zeicosatetraenoate 8peroxy5Z,9E,11Z,14Zeicosatetraenoate (s)lactate RE3288 R00703 8,9eet R00196 pyruvate carbon dioxide R00896 R03650 lcysteinyltrna(cys) lcystine sacetyldihydrolipoamide R03105 R00226 mercaptopyruvate R02050 R00621 R00268 R00709 thiamin diphosphate R00006 R00782 3carboxy1hydroxypropylthpp isocitrate 3sulfinolalanine lalanine gammalglutamyllcysteine R00369 RE2811 R00892 R02743 R02078 dopaquinone RE2642 adenosine 5triphosphate R00220 2acetolactate 7HETE adenosine 5diphosphate R00014 saminomethyldihydrolipoylprotein oxalosuccinate NacetylLcysteine R00893 RE2031 gamaLglutamylLalanine RE1915 RE2223 R03425 RE0361 lcysteine RE1473 RE1978 lseryltrna(ser) (r)4phosphopantothenoyllcysteine R04233 d4phosphopantothenate RE2265 glutathione Somatostatin fragment 314 R00497 lalanyltrna h2o2 lipoylprotein rsh R00891 R04230 R00899 trna(ala) betadfructose 6phosphate R00867 R05861 R03528 coenzyme a dfructose R00866 R00585 rsch2ch(nh3+)coo3h6no2rs choline 2aminoacrylate R00590 R01001 5aminolevulinate nh3 R00395 nacylglycine

R00875

dmannose inosine 5diphosphate R01326 R01327 inosine 5triphosphate

R00830 2oxobutanoate

fructose 3phosphate

mannose6phosphate

dalanine RE2783 dfructose 1phosphate R01330

11,12eet RE3287

12peroxy5Z,8Z,10E,14Zeicosatetraenoate 11peroxy5Z,8Z,12E,14Zeicosatetraenoate 15(R)Hydroxy(5Z,8Z,11Z,13E)eicosatetraenoate 19(s)hete R07050 12hpete R07046 R01596 15hpete R07052 R01593 RE3042 RE3452 11hydroxyeicosatetraenoate

phosphatidylethanolamine

RE3298

R01281

RE3458 RE3647 RE3649

13HETE

ethanolamine

2deoxyadenosine 5diphosphate 2deoxyadenosine 5triphosphate

5,6eet

cysgly

R01595 5hydroperoxy6trans8,11,14ciseicosatetraenoate R07041 R07048 R07054 20hete RE3289 14,15eet 16(r)hete

RE3653

R03038 R01140 5Scysteinyldopa R00876

NacetylLalanine

phosphoenolpyruvate

ltyrosyltrna(tyr)

trna(tyr)

R02918 R00731 R00734 3(4hydroxyphenyl)pyruvate

dfructose 6phosphate

RE0026 3,4dihydroxylphenylalanine R01815

R01699 R00704 (r)lactate

R00729

ltyrosine trna(cys) 4oxoglutaramate R00894 2hydroxyglutarate 5oxoproline R03534 RE3062

R00736

tyramine

RE0941

iodide

R03539

3iodoltyrosine

(s)2acetolactate RE1919

guanidinoacetate RE0159

R00565 R00270 R00214

R00258

thiocyanate 4(2aminoethyl)1,2benzenediol salsolinol 1carboxylate cyanide R00216

2methyl3oxopropanoate alphaketoglutarate (r)3amino2methylpropanoate

R04730

dihydrobiopterin ltyrosyl[alphatublin] 5,6,7,8tetrahydrobiopterin tublin

R00768 R02048 R00251 R03749 lamino acid dglucosamine 6phosphate RE1465 RE2650 R00209 R00210 arginine R00344 (5lglutamyl)lamino acid R00248 (s)malate R00243 glutamic acid R01795 tetrahydrobiopterin4acarbinolamine

peptide 3hydroxylaspartate

R00256 R00253

R04073 R00355 acetylcoa glutamine xanthosine 5phosphate R01231 2oxoglutaramate glutaminyltrna R01086 R01360 peptide laspartate oxaloacetate R00578 guanosine 5phosphate R00699 R03652 phenylpyruvate R00692 2phenylacetamide hmgcoa acetoacetylcoa nad R00227 R00235 R01357 R00410 fumarate R00408 n(larginino)succinate R00357 trna(gln) apelin (112) apelin13 succinate R00575 R00257 R00689 phenethylamine R00694 RE0938 trna(phe) lphenylalanine R01375 R00698 R03660 lphenylalanyltrna(phe)

R01082

R02164 R00412 acetoacetate R01364 aspartate 4fumarylacetoacetate acetyl adenylate R01083 R00316 R01361 citrate acetate betaalanine 3hydroxybutanoic acid hydroquinone n6(1,2dicarboxyethyl)amp lcitrulline laspartyltrna(asn) R01954 R03647 R00489 R00362

carbamoyl phosphate asparagine

R01072

deaminonad+

kinetensin 17

kinetensin 47

R01397

RE2269

RE2268

5phosphoribosylamine ncarbamoyllaspartate RE2032 lasparaginyltrna(asn) 4methyl2oxopentanoate kinetensin R03648 R01090 RE2304 5phosphoalphadribose 1diphosphate kinetensin 13

RE2644

R01135 R06893 R00317 nacetylneuraminate R01810 inosine 5phosphate R00711 R00710 nacetylbetadglucosaminylamine R03421 laspartyltrna(asp) R00488 R00526 R05577 formate

NacetylLasparagine

leucine RE2270 trna(asn)

kinetensin 48

kinetensin 18

4hydroxyphenyl acetate

acetyl phosphate

R04452 1alkylsnglycerol3phosphocholine acetaldehyde

nacetyllaspartate trna(asp) nformyllaspartate n4(acetylbetadglucosaminyl)asparagine neuromedin N (14) RE2273 lleucyltrna R01049 R03657

nacyloacetylneuraminate

1alkyl2acetylsnglycero3phosphocholine

trna(leu)

neuromedin N

ribose 5phosphate R01056 ribulose5phosphate

dxylulose 5phosphate

R01641

iminoerythrose 4phosphate R01057 R06590 R01051 R01054

dsedoheptulose 7phosphate dglyceraldehyde 3phosphate

dribose 1phosphate

aminofructose 6phosphate adpribose

dribose

3methoxy4hydroxyphenylethyleneglycol

4oxo2nonenal

acrylamide valine R05551

maltose 3methyl2oxobutanoate R01214

glucose

lauric acid betadglucose

R01520

propanoylcoa

RE2649

propanoate

R01354

propinol adenylate stearoylcoa

RE0344

stearate

dgluconic acid

RE0124 6phosphodgluconate butanoate

R01176

butanoylcoa

oxalate

malonate

(r)malate

lrhamnose aminomalonate

butanal

lnorleucine 6carboxyhexanoate phenylethanolamine 5cholestene entkaurene caprylic acid

suberic acid

decanal

aspirin

derythrose orthophosphatedgalactonate

mannitol

xylose

azulene

llyxose galacturonic acid 5oxodproline 2trans,6transfarnesol glutarate

R04881

RE3420 propenoate R03665

R05196

dodecanoyl[acylcarrier protein] RE1575

azelaic acid acid RE3515 3,4dihydroxyphenylethyleneglycol 1hydroperoxy8carboxyoctyl3(methylthio)propionic 3,4epoxynon(2E)enyl ether

lvalyltrna(val) (1,4alphadglucosyl)n

acylcarrier protein

R01521

dglucono1,5lactone

methanethiol R04880 RE3417

trna(val)

3,4epoxynonanal 3,4dihydroxymandelaldehyde

Figure 2.16 Pathway map for 7 mM glucose to 17 mM glucose obtained from Metscape.23

42

Table 2.4 Active metabolic pathways, as indicated by Metscape,23 for metabolomic analysis.

3 mM to 7 mM Aminosugars metabolism Arachidonic acid metabolism Bile acid biosynthesis Biopterin metabolism Butanoate metabolism De novo fatty acid biosynthesis Di-unsaturated fatty acid beta-oxidation Fructose and mannose metabolism Galactose metabolism Glycerophospholipid metabolism Glycine, serine, alanine and threonine metabolism Glycolysis and Gluconeogenesis Glycosphingolipid metabolism Histidine metabolism Leukotriene metabolism Linoleate metabolism Lysine metabolism Methionine and cysteine metabolism Omega-6 fatty acid metabolism Pentose phosphate pathway Phosphatidylinositol phosphate metabolism Porphyrin metabolism Propanoate metabolism Prostaglandin formation from arachidonate Purine metabolism Pyrimidine metabolism Saturated fatty acids beta-oxidation TCA cycle Tyrosine metabolism Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Valine, leucine and isoleucine degradation Vitamin B3 (nicotinate and nicotinamide) metabolism Vitamin B5 - CoA biosynthesis from pantothenate x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x -

3 mM to 17 mM x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x -

7 mM to 17 Mm x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

44

Table 2.5 List of isolated analytes as indicated by Metscape analysis.23

Analyte
(r)- malate 2-trans,6-trans-farnesol 5-cholestene 5-oxo-d-proline 6-carboxyheanoate aminomalonate aspirin azulene butanal caprylic acid decanal d-erythrose d-fructose-2-phosphate d-galactonate d-galacturonate d-ribonate diethanolamine d-ribonate elaidic acid ent-kaurene galacturonic acid glutarate hexanoate hydroxylamine l-arabinfuranose l-arabinose l-lyxose l-norleucine l-norvaline l-octanol l-rhamnose malonate mannitol myristoleic acid orthophosphate orthophosphate oxalate pentadecane suberic acid tridecane xylose

3 mM to 7 mM 3 mM to 17 mM 7 mM to 17 mM
x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

45

Discussion Metabolite Profiling Twelve more target analytes than what was detected in a previous GC/MS study were indentified in this work.13 The novel targets detected are methionine, phenylalanine, tyrosine, F6P, succinate, glutamate, leucine, isoleucine, 2PG, 3PG and lactate, all as trimethylsilyl derivatives. Additionally, in the previous GC/MS study, 13 hydroxyproline was detected but, in this study, proline was detected and identified. Compared to the previous GC/MS report, the biological RSDs reported here are slightly high for 3 mM and 7 mM glucose but the technical RSDs are in good agreement at all glucose concentrations.13 It is possible that the higher biological irreproducibility at 3 mM and 7 mM glucose is a result of fewer replicates for these glucose concentrations. Due to a series of instrument and human errors only three plates were analyzed for both 3 mM and 7 mM glucose, while all four plates were analyzed at 17 mM glucose. Consistent with previous results,13,
30

glucose, G6P, pyruvate, citrate, fumarate,

succinate and malate all increased from 3 to 7 to 17 mM glucose (see Figures 2.17 and 2.18). As illustrated in Figure 2.17, unlike the previous study where G6P is only detected at 16.8 mM glucose,13 G6P is detected at all three glucose concentrations. The lack of a statistical significance in the increase of G6P from 3 to 7 to 17 mM glucose can likely be explained by the large error bars which is was also observed at 16.8 mM in the previous data.13 Additionally, 2PG decreased from 3 mM glucose to 17 mM glucose. F6P, 3PG, R5P and lactate did not show any statistically significant change with glucose concentration.

46

3 mM

7 mM

17 mM

P e a k A re a /P e a k A re a d 6 - S u c c in a t e

0 .4

0 .3

* *

*
3

0 .2
2

0 .1

0 .0

6P

F6

at

3P

Figure 2.17 Effect of glucose on detectable glycolysis analytes. Statistical significance was tested using one-way ANOVA analysis. (*) is statistically different than 3 mM and (#) is statistically different from 7 mM.

3 mM
P e a k A re a /P e a k A re a d 6 - S u c c in a t e

yr

uv

7 mM *# *
0 .5 0 0 .7 5

17 m M

14 12 10 8 6 4 2 0

*# * *
e e at at
#

0 .2 5

0 .0 0

te

at

5P

na

i tr

ar

al

ci

Fu

Figure 2.18 Effect of glucose on detectable TCA and pentose phosphate shunt (R5P) analytes. Statistical significance was tested using one-way ANOVA analysis. (*) is statistically different than 3 mM and (#) is statistically different from 7 mM.

The changes to the amino acids detected can be seen in Figure 2.19. Consistent with previous GC/MS work 13 alanine increased and valine, glycine, and threonine did not change significantly. Although ornithine and serine do not show statistically significant differences, both analytes follow trends similar to that shown in previous work 13 with serine increasing and ornithine decreasing. The major differences between this work and 47

uc

La

ct

at

2P

3 mM
P e a k A re a /P e a k A re a d 6 - s u c c in a t e

7 mM

17 m M
0 .3

30

1 .5

*#
20 1 .0

* *

0 .2

10

0 .5

0 .1

* *
0

*
0 .0

0 .0
ne e ne e

ne

te

ne

ta

ni

al

uc

er

ni

on

ro

ci

in

ci

in

ar

ly

ta

eu

on

it h

la

la

Le

sp

la

re

rn

lu

hi

Th

ny

et

he

Is

Figure 2.19 Effect of glucose on detectable amino acids. Statistical significance was tested using one-way ANOVA analysis. (*) is statistically different than 3 mM and (#) is statistically different from 7 mM

the previous study13 are that leucine, proline, phenylalanine, isoleucine, methionine and tyrosine were detected. Additionally, in the previous study13 aspartate and glutamate decreased whereas in this work aspartate decreased from 3 mM glucose to 7 mM glucose and increased from 7 mM glucose to 17 mM and glutamate increased from 3 mM to 7 mM to 17 mM glucose. These differences may be the result of different derivatization methods, differences in the length of time that the cells were stimulated for or differences in how long the cells were allowed to equilibrate in low glucose before stimulation at higher glucose concentrations. Metabolomic Analysis For the studies reported above, we measured metabolites at 3 different glucose concentrations. The step from 3 to 7 to 17 mM glucose increases insulin secretion, so pathways activated may be involved in glucose-stimulated insulin secretion. Further, chronic stimulation with 17 mM glucose may lead to glucotoxicity in INS-1 cells;8 therefore pathways activated at this highest glucose concentration may be candidates for 48

Ty

ol

ro

si

ne

li n

at

in

in

in

in

involvement in development of glucotoxicity. Based on this consideration, we analyzed the pathways activated using MetScape.23 In addition to the distinct visual differences seen in Figures 2.14 through 2.16, the maps also differed in the pathways, seed analytes and number of reactions involved. This information can be found for all three data sets in Appendices D through F. Not surprisingly, three of the metabolic pathways involved for each data set were the citric acid cycle (TCA cycle), the pentose phosphate pathway and glycolysis and gluconeogenesis, which, as discussed previously, are known to be involved in GSIS. These data are in good agreement with the metabolite profiling data such that all of the seed analytes involved in these pathways, as seen in Appendices D through F, that overlap with previous target analytes change in the same direction. That is, malate, fumarate, citrate and pyruvate all increase from low to moderate to high glucose while ribose-5-phosphate decreases with increasing glucose in both the metabolite profiling and metabolomics data analysis. Other activated pathways with potential links to diabetes, GSIS and glucotoxicity are discussed below. Linoleate, or linoleic acid (LA), metabolism is shown in Figure 2.20.27-29 LA is an essential, omega-6 fatty acid that must be metabolized to be utilized by the body and is required for the biosynthesis of arachidonic acid (AA). Thus, as LA concentration decreases, AA concentration should increase. The increase in AA observed in this work is plotted in Figure 2.21. If LA concentration were plotted with AA, the trend would be exactly the opposite of that seen for AA. However, unlike AA, LA detection was not reproducible with RSDs of >100% and therefore was not included in Figure 2.21 for clarity. In addition to being linked to AA metabolism, LA metabolism is of interest in this 49

50
Figure 2.20 Linoleic acid metabolism pathway as obtained by KEGG database.27-29 Blue boxes indicate analytes detected in metabolomic analysis. Boxes filled from left to right indicating if the analytes increased (blue), decreased (yellow), or did not change (red) from 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose.

context because the first step in linoleate metabolism is the conversion of LA to -linoleic acid (GLA) by -6-desaturase, this is also the rate limiting step in LA metabolism. 31, 32 In 1955 it was discovered that diabetic animals require more LA than non-diabetic animals,33 a requirement that was later explained by an impairment of -6-desaturase enzyme activity and thus LA to GLA conversion. 34 It is possible that glucotoxic conditions contribute to the impairment of -6-desaturase enzyme activity and therefore it would be interesting to further investigate this pathway under such conditions.

6 Average Peak Area (x10 4)

2
0 5 10 [glucose] mM
Figure 2.21 Plot of the average peak area of arachidonic acid at 3, 7 and 17 mM glucose indicating the substantial increase of AA at 17 mM glucose. Error bars are SEM and * indicates statistical significance as calculated using a students t-test.

15

20

In our work, we found that AA varied in AA metabolism (Figure 2.22),27-29 prostaglandins metabolism and leukotriene metabolism for 7 mM to 17 mM and 3 mM to 17 mM but not 3 mM to 7 mM glucose, see Appendices D through F. Additionally, as can be seen in Figure 2.21, AA was significantly elevated at 17 mM glucose relative to the lower glucose concentrations which is consistent with a previous GC study.35

51

52
Figure 2.22 Arachidonic acid metabolism pathway as obtained from KEGG database.27-29 Blue boxes indicate analytes detected by metabolomic analysis. Boxes filled from left to right indicating if the analytes increased (blue), decreased (yellow), or did not change (red) from 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose.

In this previous study the authors stimulated islets at 3 mM and 17 mM glucose and analyzed the relative abundance of free AA as well as palmitate, linoleate, oleate and stearate by extraction and methylesterfication followed by GC/MS. 35 Arachidonic acid is an important omega-6 fatty acid that is involved in cellular signaling, the normal function of pancreatic -cells, GSIS, and can be linked to a number of diseases including obesity.36-38 Exogenous AA has been shown to enhance insulin secretion from -cells and a reduction of endogenous AA has been shown to significantly reduce GSIS in human islets.38 Recent work has also demonstrated that AA (an unsaturated fatty acid) has a positive effect on attenuating the negative effects of palmitic acid (a saturated fatty acid).38 Palmitic acid can lead to excessive generation of ROS38 , which can contribute to glucotoxicity. Thus, the recent result of AA mediated rescue of cells from palmitic acid mediated dysfunction has led to the discussion of further investigation of its metabolism and metabolites to better understand and potentially treat diabetes. 38 Although AA acid can have a protective effect, its metabolites can also have a negative effect on insulin secretion. Specifically, cyclooxygenase (COX)-generated and lipoxygenase (LOX)-generated arachidonic acid metabolites which are associated with these potential destructive effects. COX activity leads to prostaglandins and LOX produce leukotrienes.39 Both prostaglandins and leukotrienes mediate signals of inflammation39 which is an important pathological process that leads to -cell dysfunction and death in type 2 diabetes.40 Additionally, COX activity can be responsible for the production of ROS such as hydrogen peroxide39 which can further contribute to -cell dysfunction and glucotoxicity. However, it has been shown that AA metabolism through COX and LOX pathways is not required for AA to have a stimulatory effect on 53

human islets.41 Therefore, it is thought that selective inhibition of these enzymes would have a dual protective role; it would minimize -cell dysfunction and enhance endogenous arachidonic acid levels.41 Another involved pathway is butanoate metabolism (see Figure 2.2327-29) which, in Appendices D through F, shows from 3 mM to 17 mM glucose, acetoacetate and 3-hydroxybutanoic acid increase, from 7 mM to 17 mM glucose, acetoacetate increases and butanoate decreases, and from 3 mM to 7mM glucose, butanoate decreases. This decrease in butanoate is of interest because in addition to being linked to the citric acid cycle, glycolysis and the synthesis and degradation of ketone bodies as shown in Figure 2.23,27-29 dietary supplements of butanoate have recently been shown to improve insulin sensitivity in mice.42 It is suspected that butanoate stimulates mitochondrial function through the induction of peroxisome proliferator-activated receptor (PPAR)- coactivator PGC-1, which is a transcriptome activator .42 PGC-1 controls energy metabolism by interaction with several transcription factors that direct gene transcription for mitochondrial biogenesis and respiration and a reduction in the function of PGC-1 is related to reduction in fatty acid oxidation, mitochondrial dysfunction and risk for insulin resistance.42 A number of the metabolites involved in butanoate metabolism, which can be found in Figure 2.23,27-29 can be detected by the technique used in this work. Therefore it may be interesting to repeat this experiment with a butanoate treatment prior to glucose stimulation. Additionally, if this method could be adapted to islet analysis, it could be used to analyze healthy islets as well as diseased islets before and after dietary butanoate supplementation.

54

55
Figure 2.23 Butanoate metabolism pathway obtained from KEGG database.27-29 Blue boxes indicate analytes detected by metabolomics analysis, red boxes indicate analytes detected by metabolite profiling and green boxes indicate analytes detected in both methods. Boxes filled from left to right indicating if the analytes increased (blue), decreased (yellow), or did not change (red) from 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose.

As seen in Appendices D through F, serine is only increasing in glycerophospholipid metabolism at 3 mM to 17 mM and 7 mM to 17 mM glucose. This can likely be explained by the relation to the glycine, serine and threonine metabolism pathway (includes alanine in appendices) that feeds into glycerophospholipid metabolism as seen in Figure 2.24.27-29 As observed in Appendices D through F, serine is also only involved for glycine, serine and theronine metabolism at 3 mM to 17 mM and 7 mM to 17 mM glucose. The same is true for glycosphingolipid metabolism, shown in Figure 2.25,27-29 which is also fed serine by the glycine, serine, and threonine metabolism pathway. The precursor of all complex glycosphingolipids is ceramide which is formed by de novo synthesis or catabolism of glycosphingolipids and sphingomyelin. The rate of de novo synthesis is regulated by the availability of the precursors serine as well as palmitoyl-CoA.43 It is well established that glycosphingolipids are involved in intercellular communication events and cell differentiation; 43 however, several studies have indicated that glycosphingolipids interfere directly with insulin signaling and pharmacological reduction of glycosphingolipid synthesis has been found to improve insulin resistance in rodents.43 Thus, evidence is accumulating for a role for glycosphingolipids in insulin resistance. It is interesting to note that, although the increasing trend in serine concentration is apparent in Figure 2.18, the previously discussed one-way ANOVA analysis did not find a statistical significance between the concentrations at 3, 7 and 17 mM glucose and the Fisher Ratio analysis did. This highlights the advantage of using the Fisher Ratio analysis; the ability to consider all data simultaneously and objectively identify the most significant differences.21 56

57
Figure 2.24 Glycerophospholipid metabolism pathway obtained from KEGG database. 27-29 Green boxes indicate pathways containing analytes detected in both the metabolite profiling and metabolomics analysis.

58
Figure 2.25 Glycosphingolipid metabolism pathway as obtained from KEGG database. 27-29 Green boxes indicate pathways containing analytes detected in both the metabolite profiling and metabolomics analysis.

Vitamin B3 or, nicotinate and nicotinamide, (which are precursors of the coenzymes nicotinamide-adenine dinucleotide (NAD+) and nicotinamide-adenine dinucleotide phosphate (NADP+)) metabolism and vitamin B5-CoA biosynthesis from pantothenate (the main function of which is to synthesize CoA, an essential coenzyme in a variety of reactions) are two pathways that are also interesting (see Figures 2.26 and 2.27).27-29 Until recently, neither pathway had been associated with insulin secretion. A previous study looking at NMR-based metabolomic and transcriptomic data from urine, blood, liver, inguinal subcutaneous fat and skeletal muscle collected from db/db diabetic mice reported novel disease effects on nicotinamide metabolites and pantothenate. 44 A second study utilizing HPLC analysis of urine from 14 diabetic and 14 non-diabetic patients as well as metabolic data from adult Sprague-Dawley rats found that nicotinamide overload induced an increase in plasma N 1 -methylnicotinamide which is associated with oxidative stress and insulin resistance, thus playing a role in type 2 diabetes.45 Thus, it would be interesting to further investigate the role of this pathway in GSIS and -cell dysfunction.

59

60
Figure 2.26 Vitamin B3 (nicotinate and nicotinamide) metabolism as obtained from the KEGG database.27-29 Blue box indicates analytes detected by metbolomic analysis, green boxes indicate analytes detected by both metabolite profiling and metaboloic analysis. Boxes filled from left to right indicating if the analytes increased (blue), decreased (yellow), or did not change (red) from 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose.

61
Figure 2.27 Vitamin B5-CoA biosynthesis from pantothenate pathway as obtained from the KEGG database.27-29 Blue boxes indicate analytes detected by metabolomic analysis, green boxes indicate analytes (and glycolysis which contains analytes detected by) metabolite profiling and metabol omic analysis. Boxes filled from left to right indicating if the analytes increased (blue), decreased (yellow), or did not change (red) from 3 mM glucose to 7 mM glucose, 3 mM glucose to 17 mM glucose and 7 mM glucose to 17 mM glucose.

The urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine pathways (see Figure 2.2827-29), the main function of which is to excrete the excess nitrogen resulting from metabolic breakdown of amino acids, 46 shows an increase in involved analytes from 3 mM to 7 mM and 3 mM to 17 mM glucose as determined by Metscape and can been seen in Appendices D through F. These increases are particularly dramatic for fumarate, glutamine, glycine, proline and urea which increase 588, 1490, 614, 194 and 985 % respectively from 3 mM glucose to 17 mM glucose. Although no direct link to GSIS, -cell dysfunction or glucotoxicity has been previously identified, it has been hypothesized that a break down in the urea cycle may lead to the generation of pathogenic nitric oxide which may cause the destruction of -cells.47 Given this hypothesis and the involvement of the pathway observed in our study, it appears that urea cycle and metabolism arginine, proline, glutamate, aspartate and asparagine is a pathway worth focusing on in future studies. Tyrosine metabolism, the main function of which is protein synthesis as well as serving as a precursor to neurotransmitters and hormones, is also an interesting pathway. However, previous to this work, this pathway has not been directly related to insulin release, diabetes or glucotoxicity. The pathway appears to be more involved from 3 mM to 7 mM and 7 mM to 17 mM than 3 mM to 17 mM glucose, perhaps indicating an inhibition of the pathway at high glucose concentrations. This is not surprising given that the tyrosine metabolism pathway, as seen in Figure 2.29,27-29 is directly related to the citric acid cycle, another pathway which is directly affected by glucose concentration. Specifically, tyrosine can decompose into acetoacetate and fumarate, both of which can then enter the citric acid cycle. 62

63
Figure 2.28 The urea cycle and metabolism of arginine and proline as obtained from the KEGG database.27-29 Blue boxes indicate analytes detected by metabolomics analysis and green boxes indicate analytes detected by metabolite profiling and metabolmics analysis.

64
Figure 2.29 Tyrosine metabolism pathway as obtained from KEGG database.27-29 Blue boxes indicate analytes detected by metabolomics analysis, red boxes indicate analytes detected by metabolite profiling and green boxes indicate analytes detected in both methods.

Conclusion We have demonstrated the utility of GC GC-ToFMS for both metabolite profiling and metabolomic analysis of mammalian cells with excellent reproducibility. The number of detectable, target analytes has been nearly doubled compared to previous GC analysis.13 Additionally non-target analytes and the associated metabolic pathways have been investigated. Comparison of the metabolite profiling and metabolomic analysis shows good agreement between both data sets. Metabolomic analysis has demonstrated 9 activated pathways with potential or previously described relationships to GSIS, -cell dysfunction and diabetes. Metabolites from these pathways could be included in future target analysis to gain a better understanding of the effects of glucose on INS-1 cell metabolism. Further experiments could be performed to better understand the non-target analytes and the associated metabolic pathways.

65

References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Goodacre, R.; Vaidyanathan, S.; Dunn, W. B.; Harrigan, G. G.; Kell, D. B. Trends Biotechnol. 2004, 22, 245-252. Hollywood, K.; Brison, D. R.; Goodacre, R. Proteomics 2006, 6, 4716-4723. Kulkarni, R. N. The International Journal of Biochemistry & Cell Biology 2004, 36. Edwards, J. L., University of Michigan, Ann Arbor, 2006. Fernandez, C.; Fransson, U.; Hallgard, E.; Spegel, P.; Holm, C.; Krogh, M.; Warell, K.; James, P.; Mulder, H. J. Proteome Res. 2008, 7, 400-411. Straub, S. G.; Sharp, G. W. G. Diabetes-Metabolism Research and Reviews 2002, 18, 451-463. Komatsu, M.; Sato, Y.; Aizawa, T.; Hashizume, K. Endocr. J. 2001, 48, 275-288. Poitout, V.; Robertson, R. P. Endocr. Rev. 2008, 29, 351-366. Unger, R. H.; Grundy, S. Diabetologia 1985, 28, 119-121. Dubois, M.; Vacher, P.; Roger, B.; Huyghe, D.; Vandewalle, B.; Kerr-Conte, J.; Pattou, F.; Moustaid-Moussa, N.; Lang, J. C. Endocrinology 2007, 148, 16051614. Roche, E.; Farfari, S.; Witters, L. A.; Assimacopoulos-Jeannet, F.; Thumelin, S.; Brun, T.; Corkey, B. E.; Saha, A. K.; Prentki, M. Diabetes 1998, 47, 1086-1094. Nolan, C. J.; Prentki, M. Trends Endocrinol. Metab. 2008, 19, 285-291. Fernandez, C.; Fransson, U.; Hallgard, E.; Spegel, P.; Holm, C.; Krogh, M.; Warell, K.; James, P.; Mulder, H. J. Proteome Res. 2008, 7, 400-411. Hope, J. L.; Prazen, B. J.; Nilsson, E. J.; Lidstrom, M. E.; Synovec, R. E. Talanta 2005, 65, 380-388. Pierce, K. M.; Hoggard, J. C.; Hope, J. L.; Rainey, P. M.; Hoofnagle, A. N.; Jack, R. M.; Wright, B. W.; Synovec, R. E. Analytical Chemistry 2006, 78, 5068-5075. Huang, X. D.; Regnier, F. E. Anal. Chem. 2008, 80, 107-114. Li, X.; Xu, Z.; Lu, X.; Yang, X.; Yin, P.; Kong, H.; Yu, Y.; Xu, G. Anal. Chim. Acta 2009, 633, 257-262. Shellie, R. A.; Welthagen, W.; Zrostlikova, J.; Spranger, J.; Ristow, M.; Fiehn, O.; Zimmermann, R. J. Chromatogr. A 2005, 1086, 83-90. Welthagen, W.; Shellie, R. A.; Spranger, J.; Ristow, M.; Zimmermann, R.; Fiehn, O. Metabolomics 2005, 1, 65-73. Kusano, M.; Fukushima, A.; Kobayashi, M.; Hayashi, N.; Jonsson, P.; Moritz, T.; Ebana, K.; Saito, K. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences 2007, 855, 71-79. Mohler, R. E.; Dombek, K. M.; Hoggard, J. C.; Young, E. T.; Synovec, R. E. Anal. Chem. 2006, 78, 2700-2709. Mohler, R. E.; Tu, B. P.; Dombek, K. M.; Hoggard, J. C.; Young, E. T.; Synovec, R. E. J. Chromatogr. A 2008, 1186, 401-411. Gao, J.; Tarcea, V. G.; Karnovsky, A.; Mirel, B. R.; Weymouth, T. E.; Beecher, C. W.; Cavalcoli, J. D.; Athey, B. D.; Omenn, G. S.; Burant, C. F.; Jagadish, H. V. Bioinformatics, 26, 971-973. Cline, M. S.; Smoot, M.; Cerami, E.; Kuchinsky, A.; Landys, N.; Workman, C.; Christmas, R.; Avila-Campilo, I.; Creech, M.; Gross, B.; Hanspers, K.; Isserlin, 66

(11) (12) (13) (14) (15) (16) (17) (18) (19) (20)

(21) (22) (23)

(24)

(25) (26) (27) (28) (29)

(30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41)

(42) (43) (44) (45)

(46) (47)

R.; Kelley, R.; Killcoyne, S.; Lotia, S.; Maere, S.; Morris, J.; Ono, K.; Pavlovic, V.; Pico, A. R.; Vailaya, A.; Wang, P. L.; Adler, A.; Conklin, B. R.; Hood, L.; Kuiper, M.; Sander, C.; Schmulevich, I.; Schwikowski, B.; Warner, G. J.; Ideker, T.; Bader, G. D. Nature Protocols 2007, 2, 2366-2382. Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N. S.; Wang, J. T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Genome Res. 2003, 13, 2498-2504. Mohler, R. E.; Dombek, K. M.; Hoggard, J. C.; Pierce, K. M.; Young, E. T.; Synovec, R. E. Analyst 2007, 132, 756-767. Kanehisa, M.; Goto, S. Nucleic Acids Res. 2000, 28, 27-30. Kanehisa, M.; Goto, S.; Furumichi, M.; Tanabe, M.; Hirakawa, M. Nucleic Acids Res. 2010, 38, D355-D360. Kanehisa, M.; Goto, S.; Hattori, M.; Aoki-Kinoshita, K. F.; Itoh, M.; Kawashima, S.; Katayama, T.; Araki, M.; Hirakawa, M. Nucleic Acids Res. 2006, 34, D354D357. Alarcon, C.; Barton, W.; Prentki, M.; Corkey, B. E.; Rhodes, C. J. Diabetes 2002, 51, 2496-2504. Horrobin, D. F. Prog. Lipid Res. 1992, 31, 163-194. Horrobin, D. F. Am. J. Clin. Nutr. 1993, 57, S732-S737. Peifer, J. J.; Holman, R. T. Arch. Biochem. Biophys. 1955, 57, 520-521. Mercuri, O.; Peluffo, R. O.; Brenner, R. R. Biochim. Biophys. Acta 1966, 116, 409-&. Wolf, B. A.; Pasquale, S. M.; Turk, J. Biochemistry 1991, 30, 6372-6379. Wymann, M. P.; Schneiter, R. Nat Rev Mol Cell Biol 2008, 9, 162-176. Ramanadham, S.; Hsu, F.-F.; Zhang, S.; Bohrer, A.; Ma, Z.; Turk, J. BBA-Mol. Cell. Biol. L. 2000, 1484, 251-266. Keane, D.; Takahashi, H. K.; Dhayal, D.; Morgan, N. G.; Curi, R.; Newsholme, P. Clin. Sci. 2010, Immediate Publication. Meves, H. Br. J. Pharmacol. 2008, 155, 4-16. Ma, K.; Nunemaker, C. S.; Wu, R.; Chakrabarti, S. K.; Taylor-Fishwick, D. A.; Nadler, J. L. J. Clin. Endocrinol. Metab., 95, 887-893. Persaud, S. J.; Muller, D.; Belin, V. D.; Kitsou-Mylona, I.; Asare-Anane, H.; Papadimitriou, A.; Burns, C. J.; Huang, G. C.; Amiel, S. A.; Jones, P. M. Diabetes 2007, 56, 197-203. Gao, Z. G.; Yin, J.; Zhang, J.; Ward, R. E.; Martin, R. J.; Lefevre, M.; Cefalu, W. T.; Ye, J. P. Diabetes 2009, 58, 1509-1517. Langeveld, M.; Aerts, J. Prog. Lipid Res. 2009, 48, 196-205. Connor, S. C.; Hansen, M. K.; Corner, A.; Smith, R. F.; Ryan, T. E. Mol. BioSyst. 2010, 6, 909-921. Zhou, S. S.; Li, D.; Sun, W. P.; Guo, M.; Lun, Y. Z.; Zhou, Y. M.; Xiao, F. C.; Jing, L. X.; Sun, S. X.; Zhang, L. B.; Luo, N.; Bian, F. N.; Zou, W.; Dong, L. B.; Zhao, Z. G.; Li, S. F.; Gong, X. J.; Yu, Z. G.; Sun, C. B.; Zheng, C. L.; Jiang, D. J.; Li, Z. N. World Journal of Gastroenterology 2009, 15, 5674-5684. Voet, D.; Voet, J. G. Biochemistry, 3 ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2004. Bauer, J. A. Med. Hypotheses 1998, 51, 71-73.

67

Chapter 3 ANALYSIS OF LIPID COMPOSITION IN INS-1 CELLS VIA COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAPHY TIME-OF-FLIGHT MASS SPECTROMETRY Introduction Fatty acids play an important role in nutrition, health, and biochemistry. The consumption of fatty acids has been linked to a number of diseases including heart disease, hypertension, diabetes, rheumatoid arthritis and Chrohns disease.1-3 Fatty acids also serve as biomarkers; for example, docosahexaenoic acid has been used as an indicator of severe left ventricular dilation in heart failure patients.4 Fatty acids can also serve as signaling molecules in a variety of ways including activation of receptors such as PPAR
5

Furthermore, imbalances in lipid signaling pathways have been shown to

contribute to the progression of chronic inflammation, autoimmunity, cancer, hypertension, and metabolic diseases.6 As a result of the significance of fatty acids, their measurement in biological samples is of interest. Fatty acid analysis is complicated by their wide variety found in biological samples. They occur with a wide range of carbon chain lengths that may be saturated, monounsaturated, and polyunsaturated. Fatty acids may be free or bound as esters in lipids such as triacylglycerols, glycerophospholipids, and sphingolipids. Fatty acids are commonly analyzed via gas chromatography (GC) after transformation into methyl esters. Other methods such as pyrolysis GC
7

and high

performance liquid chromatography (HPLC) have also been applied for fatty acid 68

measurements.8 Multidimensional techniques including comprehensively coupled liquid chromatography and gas chromatography (LC GC)9,10 and comprehensive two-dimensional gas chromatography (GC GC) are also gaining popularity.11, 12 It has been shown that the use of GC GC instead of traditional GC to determine fatty acid methyl esters (FAMEs) enhances overall resolution, which is important because these samples often contain compounds with similar structures that are difficult to separate in a one-dimensional analysis.3, 13 In GC GC, FAMEs with the same carbon number elute as clusters with predictable patterns. As carbon number increases, first dimension retention time increases so that a homologous series appears along an arc in a traditional contour plot. Esters with the same number of double bonds appear on arcs offset from that for saturated fatty acids. The inherent structure of these chromatograms aids in identification of FAMEs even when a pure standard is not available. Improved resolution and detectability, compared to one-dimensional GC, also make it possible to detect and identify odd chain fatty acids that are usually present in trace amounts.11, 13 GC GC has previously been used to analyze fatty acids in biodiesel, 14, 15 human plasma,3 algae,16 and edible fats and oils including subcutaneous pig, cow, sheep, and poultry tissue as well as vegetable, fish, peanut, and olive oils.11-13,
17

Despite this

extensive list of applications for GC GC analysis of fatty acids, little work has focused on developing or evaluating methods for analysis of fatty acids in cell culture lines by GC GC. Such analysis is likely to be of value in understanding the role of fatty acids in signaling and disease pathophysiology. In this work, we demonstrate the applicability of GC GC for determination of total fatty acid content in the INS-1 cell culture line.

69

INS-1 cells are a -cell line and useful model for insulin secretion and diabetes studies. In -cells, fatty acids may serve as signaling molecules or energy sources.6, 18, 19 Disturbances of fatty acid metabolism have been implicated in diabetes.6,
18

Therefore,

determination of fatty acid content may help elucidate the roles of fatty acids in these cells and their link to insulin secretion and diabetes. Previous methods for determining fatty acids in INS-1 cells include HPLC 8 and GC.20 With these methods, 11 and 15 fatty acids were identified and quantified, respectively. Using the GC GC method reported here, we discover an additional 15 fatty acids, most at higher carbon number and degree of saturation than previously reported, in these cells. These results establish a method for aiding studies of the role of fatty acids in -cells. These also demonstrate the utility of GC GC for the determination of FAMEs in cell culture lines, suggesting a potential role for studying signaling in other cell lines as well. Experimental Chemicals All chemicals were purchased from Sigma-Aldrich (St.Louis, MO) unless otherwise noted. Roswell Park Memorial Institute (RPMI) media, fetal bovine serum (FBS), HEPES, and penicillin-streptomycin were purchased from Invitrogen Corp (Carlsbad, CA). Cells lifters and 10 cm polystyrene non-pyrogenic culture dishes were purchased from Corning (Lowell, MA). Fifteen mL polypropene sterile centrifuge tubes, HPLC grade methanol, ACS certified sulfuric acid and potassium hydroxide were obtained from Fisher Scientific (Fairfield, NJ). Hexane was from Acros Organics (Morris Plain, NJ) and the 37 component FAME mix was from Supelco (Bellefonte, PA).

70

Samples INS-1 cells were cultured on 10 cm plates in RPMI-1640 (+l-glutamine) supplemented with 10 % FBS, 1 mM pyruvate, 10 mM HEPES, 50 M 2- -mercaptoethanol, and 1 unit penicillin-streptomycin. INS-1 cells were grown to confluence (~4 x 107 cells) in 10 cm polystyrene dishes with RPMI culture media. All cells used in a particular experiment were seeded at the same time taking care to minimize variability by using precise volumes of reagents and seed cells. Krebs-Ringer-HEPES buffer (KRHB) was prepared to contain 20 mM HEPES, 118 mM NaCl, 5.4 mM KCl, 2.4 mM CaCl, 1.2 mM MgSO4, and 1.2 mM KH2PO4 and adjusted to pH 7.4 with HCl. Cells were washed once with 10 mL of KRHB prior to treatment in 10 mL of KRHB containing 0, 0.5, 10, or 20 mM glucose for 1 h at 37 C. Each concentration was prepared in triplicate. After treatment, cells were washed once with 10 mL milli-Q water and snap frozen with liquid nitrogen. Plates were stored at -80 C until extraction. Fatty Acid Methyl Ester Synthesis Fatty acid methyl esters were synthesized from the total lipid content as described by OFallon et al.21 The only modification was that samples were not homogenized in a coffee grinder. Briefly, 5.3 mL of methanol was measured and an aliquot added to the cell plate. The plate was scraped and transferred to a 15 mL centrifuge tube. This procedure was repeated with the remainder of the methanol. Seven hundred L of 10 M potassium hydroxide was added followed by incubation in a 55 C water bath for 1.5 h with vigorous shaking every 20 min. The samples were cooled to below room

71

temperature with a cool water bath followed by the addition of 580 L of 12 M sulfuric acid and an additional 1.5 h of incubation at 55 C with shaking every 20 min. The tubes were cooled again to below room temperature with a cool water bath and 3 mL of hexane was added to each. The samples were vortexed for 5 min followed by centrifugation at 1600 g for 5 min. The hexane layer was transferred to a capped 4 dram vial and stored at -20 C until analysis. Comprehensive Spectrometry GC GC analysis was performed on a Leco Pegasus III with 4D upgrade (St. Joseph, MI). The primary column was a 30 m Rxi-1ms (0.25 mm i.d., 0.18 m film), and the secondary column was a 2 m Rtx-17silms (0.1 mm i.d., 0.1 m film) both from Restek Corporation (Bellefonte, PA). A 1 L injection was made with an Agilent 7683 automatic liquid sampler (Palo Alto, CA) in splitless mode, and four replicates were completed for each sample. For temperature programming, the primary oven was maintained at 40 C for two minutes and then increased at a rate of 30 C per minute to 160 C, the rate was then slowed to 2 C per minute until 260 C was reached and maintained for 0.5 minutes. The secondary oven and the thermal modulator were offset from the primary oven by 5 C and 30 C respectively. A modulation period, or second dimension injection frequency, of 7 s was used. The hot pulse time (length of time the hot jet fires to initiate injection on the second dimension) was 0.8 s. A flow rate of 2 mL/min ultra high purity helium with an inlet and mass spectral transfer line temperature of 250 C were maintained. A mass range of m/z 45 to 650 was collected at a rate of 200 spectra/s after a 400 s solvent delay. The ion source was maintained at 200 C. 72 Two-Dimensional Gas Chromatography Time-of-Flight Mass

The Supelco 37 component FAME mix was analyzed at 0.25, 0.75 and 2.5 mg/ml, in triplicate, under identical conditions to the cell extracts for validation of the method linearity. Additionally, the mix was analyzed neat with the following chromatographic differences. A 150:1 split ratio was used in addition to a 1 mL/min flow rate. The final temperature was 270 C and was maintained for 5 minutes. The mass spectral solvent delay was 300 s. Data Analysis Leco ChromaTOF version 4.22 was used for instrument control and data processing. The National Institutes of Standards and Technology (NIST) mass spectral library (version 2.0) was used to aid in peak identification. Statistical significance was determined in GraphPad Prism version 3.03 (La Jolla, CA) using a one way ANOVA analysis and a Newman-Keuls post-hoc test. Results and Discussion Analysis of 37 Component FAME Standard In initial experiments, a standard mixture of 37 FAMEs was analyzed to ensure that the column set and conditions used would provide adequate resolution. As shown in Figure 3.1, good resolution and ordered retention is achieved with these conditions. Clustered elution of FAMEs with the same carbon number but a varying degree of saturation is readily observed in the C18 and C20 range of the standard, for example. The elution of FAMEs with the same degree of saturation along a horizontal axis is highlighted by the colored lines. The added benefit of being able to resolve FAMEs with the same degree of saturation but different numbers can be seen in Figure 3.1 where

73

both -linolenic acid (C18:3 3) and -linolenic acid (C18:3 6) as well as cis-11,14,17eicosatrienoic acid (C20:3 3) and cis-8,11,14-eicosatrienoic acid (C20:3 6) are fully resolved. The additional advantage of resolution and detection of odd numbered FAMEs, as previously described,3, 11 is also demonstrated in Figure 3.1 where C11, C15, C17, C21 and C23 compounds are all easily detected and identified.

Retention Time (s) 2 4 6

n= 6 n= 5 n= 4 n= 3 n= 2 n= 1 n= 0 C10 C8

C12 C11

C13

C14

C C17 C15 16

C18

C22 C20 C21

C23 C24

C4

C6

300

1300 2300 3300 Retention Time (s)

Figure 3.1 GC GC chromatogram of neat 37 component FAMEs mix where n is the number of double bonds

These results validate the column set and chromatographic conditions for FAMEs used in this work. Changes from the initial methods were made in the flow rate, split ratio, and final temperature, as described in the experimental section, for faster analysis of actual samples containing fewer and less concentrated FAMEs than the commercially available standard.

74

Analysis of INS-1 Cell extracts Extracts of INS-1 cells incubated in 10 mM glucose were analyzed via GC GC after transformation of the total fatty acid concentration to methyl esters. representative chromatogram is shown in Figure 3.2. A

75

(iv)

(v)

5.3

(ii)
(3)

(iii)

3.3

Retention Time (s)

(2) (i)

(4)

(5)

1.3

TIC

603
2.4

1401
4.5
(i)

2346
(6) (7)

3060
(ii)

(1)

2.2

4.0

(8)

(9)

m/z 74 + 69

m/z 93

756
(iii)

856
(23)

1814
(iv)

1877
(v) (30)
(29)

5.2

(16)
(15)

5.9

(22) (21)
(20)

6.1

(14)

(28)
(19) (18) (17)

4.4

(11)

4.9

(12) (10)

5.0

(13)

(27) (26)

TIC

TIC

(25) (24) m/z 91 + 80 + 74

2234

2388

2661

2843

3060

3242

Retention Time (s)

Figure 3.2 GCGC total ion chromatogram (TIC) of a representative INS-1 cell extract (top). Peaks identified include lauric acid (1), myristic acid (2), palmitoleic acid (3), palmitic acid (4), stearic acid (5), linolenic acid (6), linoliec acid(7), oleic acid (8), elaidic acid (9), arachidic acid (10), cis-11-eicosenoic acid (11), cis-13-eicosenoic acid (12), eicosadienoic acid (13), cis-11,14,17-eicosatrienoic acid (14), cis-8,11,14-eicosatrienoic acid (15), arachidonic acid (16), behenic acid (17), erucic acid (18), brassic acid (19), eicosatetraenoic acid (20), eicosapentaenoic acid (21), clupanodonic acid (22), docosahexaenoic acid (23), lignoceric acid (24), nervonic acid (25), tetracosadienoic acid (26), tetracosatrienoic acid (27), tetracosatetraenoic acid (28), tetracosapentaenoic acid (29), tetracosahexaenoic acid (30). Boxes (i) through (v) are zoomed in portions of the corresponding broken boxes in the top chromatogram at either TIC, m/z 74 + 69, m/z 93, or m/z 91 + 80 + 74, as indicated

76

Peaks for myristic acid (C14:0), palmitic acid (C16:0), palmitoleic acid (C16:1). and stearic acid (C18:0) dominate the chromatogram; however, signals for other analytes are made apparent by expanding the scale or plotting select mass channels (Figure 3.2i through v). The advantages of using GC GC are highlighted in Figures 3.2ii through 3.2v. In these portions, the resolution of oleic and elaidic acids, the resolution of cis-11-eicosonic acid and cis-13-eicosonic acid, and the resolution of 3 and 6 eicosatrienoic acid, which would not be resolved in one-dimension, can be seen. In these samples, 30 FAMEs were identified with 12 to 24 carbons and 0 to 6 double bonds. Importantly, 15 fatty acids, of higher carbon number and number of double bonds, not previously identified by GC analysis of lipids in INS-1 cells,20 are detected and identified. Of the 30 FAMEs identified, 11 were not available as standards. Identification of these 11 analytes was performed by utilizing the chromatogram structure, NIST 02 mass spectral database matches and manual inspection of the mass spectra. The new lipids identified in the INS-1 cells are: lauric acid, linolenic acid, eicosadienoic acid, eicosatrienoic acid, erucic acid, brassic acid, eicosatetraenoic acid, eicosapentaenoic tetracosatrienoic acid, acid, clupanodonic and acid, nervonic acid, acid, tetracosenoic acid, acid,

tetracosatetraenoic

tetracosapentaenoic

tetracosahexaenoic acid. To evaluate the potential of the method for quantitative analysis, calibration curves were created for 14 FAMEs. Linear correlation coefficients of 0.99 or greater were achieved in the concentration range of 0.25 to 2.5 mg/ml. A representative curve for myristic acid is shown in Figure 3.3.

77

20

Peak Area (106)

15 10 5 0 0.0 1.0

R = 0.9998

2.0

3.0

Concentration (mg/ml)
Figure 3.3 Calibration curve for myristic acid. Samples were analyzed in triplicate and error bar is + 1 SEM

The average RSD for replicate injections of the same sample was 8.4% with a range of 3.14 to 23.21%, see Table 3.1. Replicate analysis of biological samples yielded average RSDs of 12.4% with a range of 4.58 to 36.1% for all analytes except for palmitic acid (0.5 and 10 mM glucose), elaidic acid (0.5 mM glucose), arachidic acid (10 mM glucose), and erucic acid (0 and 0.5 mM glucose), (Table 3.2). The reason for this variability is not well understood. Overall, reproducibility may be improved by use of an internal standard. Possible internal standards include odd chain fatty acids or using a sample prepared with deuterated methanol as an internal standard for each analyte.22 In summary, the linearity, detection limit, and reproducibility suggest that the method can be used for semi-quantitative analysis for most analytes. Use of internal standards would be expected to improve the quantification ability of the method.

78

Table 3.1 Average technical variability presented as relative standard deviations (RSDs) at 0 mM, 0.5 mM, 10 mM and 20 mM glucose. Three plates per concentration, four replicates per plate

Fatty Acid C12:0 C14:0 C16:0 C16:1 C18:0 C18:1 (9, cis) C18:1 (9, trans) C18:2 C18:3 C20:0 C20:1 (cis-11) C20:1 (cis-13) C20:2 C20:3 (3) C20:3 (6) C20:4 C22:0 C22:1 C22:2 C22:3 C22:4 C22:5 C22:6 C24:0 C24:1 C24:2 C24:3 C24:4 C24:5 C24:6

0 mM 4.7 4.6 9.2 7.2 7.7 5.6 14 6.9 5.2 6.3 12 6.8 7.9 9.2 5.8 6.7 5.6 11 20 4.3 5.4 5.9 6.6 9.7 10 20 20 8.7 9.5 6.9

0.5 mM 3.1 3.6 4.6 5.5 5.4 3.7 5.9 5.8 7.1 5.6 9.2 6.1 8.7 7.5 5.1 5.1 5.8 6.4 20 6.9 10 7.9 11 9.1 5.7 13 12 3.5 5.1 6.5

10 mM 9.2 7.1 6.4 12 5.4 6.6 10 5.6 4.1 5.6 7.6 5.6 7.6 6.5 6.4 5.5 7.1 11 9.8 7.7 9.1 7.9 14 9.2 7.3 14 17 6.4 11 8.2

20 mM 6.6 8.6 11 8.8 5.7 10 13 8.3 7.7 7.0 8.0 7.2 6.8 8.2 8.9 6.7 6.2 7.2 7.4 6.9 12 7.6 6.1 9.1 23 14 15 15 8.0 12

Average 5.9 5. 7.9 8.3 6.1 6.4 11 6.6 6.0 6.1 9.23 6.4 7.8 7.8 6.6 6.0 6.2 9.0 14 6.4 9.1 7.3 9.2 9.3 12 15 16 8.4 8.4 8.3

79

Table 3.2 Biological variability presented as relative standard deviations (RSDs) at 0 mM, 0.5 mM, 10 mM and 20 mM glucose. Three plates per concentration, four replicates per plate

Fatty Acid C12:0 C14:0 C16:0 C16:1 C18:0 C18:1 (9, cis) C18:1 (9, trans) C18:2 C18:3 C20:0 C20:1 (cis-11) C20:1 (cis-13) C20:2 C20:3 (3) C20:3 (6) C20:4 C22:0 C22:1 C22:2 C22:3 C22:4 C22:5 C22:6 C24:0 C24:1 C24:2 C24:3 C24:4 C24:5 C24:6

0 mM 12 12 16 8.2 12 8.2 21 7.9 7.1 6.3 12 8.0 10 9.5 5.5 8.7 5.5 66 20 4.9 5.4 5.8 6.4 8.2 6.1 21 16 6.6 5.8 11

0.5 mM 9.5 6.5 4.8 52 5.7 4.6 42 7.5 9.5 12 16 5.5 11 5.8 9.7 6.0 12 53 24 6.8 11 8.3 12 9.5 6.3 13 9.8 5.6 4.8 6.1

10 mM 12 7.3 8.8 44 7.2 6.7 13 57 7.1 7.6 7.8 6.4 9.6 9.6 8.4 7.2 15 17 26 10 10 8.7 17 20 14 17 18 6.5 12 10

20 mM 12 15 18 14 17 12 24 13 16 19 19 17 14 13 17 13 19 36 15 18 16 13 14 23 16 24 22 27 18 19

Average 11 10 12 30 10 7.9 25 22 9.9 11 14 9.3 11 9.5 10 8.9 13 43 21 10 11 8.9 12 15 11 19 17 11 9.9 11

Glucose Stimulated Changes in Fatty Acids Glucose evokes a number of metabolic changes in -cells that are linked to cell function. Most importantly, glucose stimulates insulin secretion through metabolism. Although some of the metabolic changes associated with glucose that evoke insulin secretion are known (e.g., an increase the ATP/ADP ratio), it is believed that as yet 80

unidentified changes also contribute to insulin secretion.

Glucose may also be an

important signal for adaptation (such as cell growth or increased insulin synthesis). Excessive glucose is known to result in glucotoxicity and impaired -cell function, which likely plays a role in type 2 diabetes. In view of the important role of glucose for -cell metabolism, we evaluated the acute effect of incubation in different concentrations of glucose on fatty acid content of INS-1 cells. Glucose concentrations of 0, 0.5, 10, and 20 mM were tested in triplicate and extracts from each were analyzed in quadruplicate. Table 3.3 summarizes the fatty acid content as a function of extracellular glucose concentration. Several fatty acids showed increases with glucose until the highest

concentration where a decrease was observed resulting in an inverted U-shaped dose-response curve (see Figure 3.4). Another group of fatty acids that included lauric, myristic, linolenic, cis-13-eicosenoic, eicosadienoic, cis-11,14,17-eicosatrinoic,

cis-8,11,14-eicosatrienoic, arachidonic, eicosapentaenoic, clupanodonic, docosaexaenoic, tetracosadienoic, tetracosatetraenoic, tetracosapentaenoic, and tetracosahexanoic acids experienced no significant change in peak area at 0, 0.5, and 10 mM glucose but showed a significant decrease at 20 mM glucose (see Table 3.3 and Figure 3.4).

81

Table 3.3 Determination of fatty acids in INS-1 cells incubated for 60 min at different glucose concentrations. Average peak area and 1 SEM at 0 mM, 0.5 mM, 10 mM and 20 mM glucose. a = significantly different from 0 mM glucose, b = significantly different from 0.5 mM glucose, c = significantly different from 10 mM glucose. Significance was determined with one way ANOVA test and Newman-Keuls post-hoc test. Three plates per concentration, four replicates per plate.

FAME C12:0 C14:0 C16:0 C16:1 C18:0 C18:1 (9, cis) C18:1 (9, trans) C18:2 C18:3 C20:0 C20:1 (cis-11) C20:1 (cis-13) C20:2 C20:3 (3) C20:3 (6) C20:4 C22:0 C22:1 C22:2 C22:3 C22:4 C22:5 C22:6 C24:0 C24:1 C24:2 C24:3 C24:4 C24:5 C24:6

Peak Area 0 mM glucose 1.1E 0.04 20 0.8 220 1 7.8 0.2 95 4 20 0.6 8.0 0.6 3.4 0.1 0.16 0.004 2.6 0.06 1.7 0.07 1.1 0.03 0.9 0.03 1.7 0.06 1.2 0.02 7.4 0.2 0.80 0.02 0.30 0.07 0.35 0.02 0.31 0.05 0.50 0.01 1.1 0.02 5.4 0.1 0.40 0.01 0.091 0.002 0.073 0.005 0.048 0.003 0.13 0.003 0.70 0.01 0.7 7 0.03

Peak Area 0.5 mM glucose 0.99 0.04 20 0.5 240 5 4.6 1.0 a 110 3 6.5 1 3.3 0.1 0.15 0.006 2.8 0.01 1.8 0.1 1.1 0.03 0.93 0.04 1.7 0.04 1.2 0.05 7.1 0.2 0.97 0.05 a 0.24 0.05 0.38 0.04 0.32 0.09 0.48 0.02 1.1 0.04 5.3 0.2 0.47 0.02 0.11 0.003 a 0.088 0.005 0.051 0.002 0.14 0.003 0.69 0.01 0.79 0.02
a

Peak Area 10 mM glucose 0.99 0.04 22 0.5 250 6 5.7 0.7 a 110 2
a

Peak Area 20 mM glucose 0.79 0.03 a,b,c 17 0.7 a,b,c 210 11 c 6.6 0.3 88 4 b,c 18 0.6 a,c 7.4 0.5 c 2.6 0.1 0.13 0.006 a,b,c 2.3 0.01 a,b,c 1.5 0.08b,c 0.93 0.05 a,b,c 0.76 0.03 a,b,c 1.3 0.05 a,b,c 1.1 0.05 a,b,c 6.0 0.2 a,b,c 0.775 0.04 b,c 0.22 0.02 c 0.32 0.01 0.25 0.01 0.40 0.02 a,b,c 0.91 0.03 a,b,c 4.5 0.2 a,b,c 0.35 0.02 b,c 0.097 0.004 0.063 0.004 b,c 0.040 0.003 0.10 0.008 a,b,c 0.59 0.03 a,b,c 0.63 0.03 a,b,c

20 0.4

20 0.04 9.6 0.4 a 2.5 0.4 0.17 0.003 3.0 0.07 c 1.9 0.04 1.2 0.02 0.97 0.03 1.7 0.05 1.3 0.03 7.5 0.2 1.0 0.04 a 0.43 0.02 a,b 0.35 0.03 0.33 0.1 0.51 0.02 1.1 0.03 5.4 0.3 0.47 0.03 0.11 0.004 a 0.082 0.004 0.47 0.002 0.14 0.003 0.70 0.02 0.81 0.02

82

Figure 3.4 Average area of palmitic acid (C16:0), stearic acid (C18:0), eicosenoic acid (C20:1), arachidonic acid (C20:4), behenic acid (C22:0) and erucic acid (C22:1) at 0 mM, 0.5 mM, 10 mM, and 20 mM glucose. Error bars are + 1 SEM, a = significantly different from 0 mM glucose, b = significantly different from 0.5 mM glucose, c = significantly different from 10 mM glucose. Significance was determined with one way ANOVA test and Newman-Keuls post hoc test

83

The significance of these changes to cell physiology cannot be discerned without further experimentation; however, these changes do illustrate the potential of the method to provide detailed analysis of the fatty acid profile. Arachidonic acid is of particular interest as it has previously been directly linked to the function of pancreatic -cells, is involved in glucose-stimulated insulin release, and can be linked to a number of diseases including obesity.6, 23 Although direct links are less obvious for other fatty acids, it has been established that fatty acids promote insulin secretion and that not all fatty acids have the same insulinotropic potency.24,
25

Previous work has shown that insulinotropic

potency is directly related to fatty acid chain length and degree of saturation with, for example, stearic acid (C18:0) having more of an effect on insulin secretion than palmitic acid (C16:0) and oleic acid (C18:1).25 As glucose induces changes in metabolism as

well as insulin secretion, such changes may be involved in both these processes. Further, excessive glucose can be toxic to INS-1 cells, so the changes that occur at 20 mM glucose may be of interest for understanding the mechanisms of glucotoxicity. Many fatty acids showed little change as a function of glucose concentration. Because we measured total fatty acids, these stable results do not preclude possible redistribution between different pools, e.g. free and esterfied fatty acids. For example, a previous study has shown substantial changes in myristic acid, palmitic acid, palmitoleic acid, stearic acid, oleic acid, and vaccenic acid in the phospholipid fraction of INS-1 cell extracts,20 and only minor changes in free fatty acids or other lipid classes. Our results do not completely agree with a previous study that measured 11 fatty acids (total fatty acid content) in pancreatic islets of Langerhans (microorgans containing -cells) as in the presence of 0, 5.6, 8.3 and 16 mM glucose via HPLC.8 In agreement 84

with this previous study, we observed an increase in stearic acid with glucose concentration and no statistically significant difference for linoleic acid. At moderate glucose concentrations, arachidonic and docosahexaenoic acid concentrations are not statistically different in both the work presented here and the previous islet study.8 However, at high glucose, we observe a decrease in arachidonic and docosahexaenoic acids that was not previously observed in islets.8 Additionally, an increase in lauric acid (C12:0), myristic acid (C14:0), palmitic acid (C16:0), and palmitoleic acid (C16:1) concentrations when compared to the absence of glucose were reported in the previous work
8

with islets but not in the INS-1cell data presented here. The changes observed

previously8 in lauric acid and myristic acid are only seen at 16 mM glucose, thus, in view of the inverted U-shaped curve observed in our work, it is possible that these changes are missed in our experiment. The difference in palmitic and palmitoleic acid results may be explained by the high variability observed for these analytes. Another important point is that although islets and INS-1 cells are metabolically similar; they also have substantial differences that could easily affect the lipid profile and content. Conclusions We have demonstrated that GC GC is a viable tool for lipid analysis not only in fats and oils but also cultured cells. The utility of this technique is highlighted by the identification of 15 fatty acid methyl esters not identified in previous analyses of these cells.8,
20

The potential of the method was also shown for monitoring lipid changes This result hinged on the

following perturbations, such as glucose concentration.

excellent reproducibility achieved for most of the fatty acids. Further analysis using an internal standard and/or separating lipid classes prior to FAMEs analysis could reveal 85

valuable information about INS-1 cell metabolism. Additionally, applying the methodology developed in this work to different cell lines may help further elucidate the role of fatty acids in human health and disease states.

86

References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) Simopoulos, A. P. The American Journal of Clinical Nutrition 1999, 70. Seppnen-Laakso, T.; Laakso, I.; Hiltunen, R. Anal. Chim. Acta 2002, 465, 39-62. Quinto Tranchida, P.; Costa, R.; Donato, P.; Sciarrone, D.; Ragonese, C.; Dugo, P.; Dugo, G.; Mondello, L. J. Sep. Sci. 2008, 31, 3347-3351. Rupp, H.; Rupp, T.; Alter, P.; Maisch, B. Br. Heart J. 2010, 96, 595-598. Rokling-Andersen, M.; Rustan, A.; Wensaas, A.; Kaalhus, O.; Wergedahl, H.; Rost, T. Br. J. Nutr. 2009, 102, 995-1006. Wymann, M. P.; Schneiter, R. Nat Rev Mol Cell Biol 2008, 9, 162-176. Fabbri, D.; Baravelli, V.; Chiavari, G.; Prati, S. J. Chromatogr. A 2005, 1100, 218-222. Martins, E. F.; Miyasaka, C. K.; Newsholme, P.; Curi, R.; Carpinelli, A. R. Diabetes Metab. 2004, 30, 21-27. Janssen, H.-G.; Boers, W.; Steenbergen, H.; Horsten, R.; Flter, E. J. Chromatogr. A 2003, 1000, 385-400. Koning, S. d.; Janssen, H.-G.; Deursen, M. v.; Brinkman, U. A. T. J. Sep. Sci. 2004, 27, 397-409. Adahchour, M.; Beens, J.; Brinkman, U. A. T. J. Chromatogr. A 2008, 1186, 67108. Chin, S.-T.; Che Man, Y. B.; Tan, C. P.; Hashim, D. M. J. Am. Oil Chem. Soc. 2009, 86, 949-958. Dalluge, J.; Beens, J.; Brinkman, U. A. T. J. Chromatogr. A 2003, 1000, 69-108. Tiyapongpattana, W.; Wilairat, P.; Marriott, P. J. J. Sep. Sci. 2008, 31, 26402649. Adam, F.; Bertoncini, F.; Coupard, V.; Charon, N.; Thiebaut, D.; Espinat, D.; Hennion, M. J. Chromatogr. A 2008, 1186, 236-244. Akoto, L.; Stellaard, F.; Irth, H.; Vreuls, R.; Pel, R. J. Chromatogr. A 2008, 1186, 254-261. Mondello, L.; Casilli, A.; Tranchida, P. Q.; Dugo, P.; Dugo, G. J. Chromatogr. A 2003, 1019, 187-196. Yaney, G. C.; Corkey, B. E. Diabetologia 2003, 46, 1297. Nolan, C. J.; Madiraju, M. S. R.; Delghingaro-Augusto, V.; Peyot, M.-L.; Prentki, M. Diabetes 2006, 55, S16-S23. MacDonald, M. J.; Dobrzyn, A.; Ntambi, J.; Stoker, S. W. Arch. Biochem. Biophys. 2008, 470, 153-162. O'Fallon, J. V.; Busboom, J. R.; Nelson, M. L.; Gaskins, C. T. J. Anim. Sci. 2007, 85, 279-280. Li, J.; Yue, Y.; Hu, X.; Zhong, H. Anal. Chem. 2009, 81, 5080. Ramanadham, S.; Hsu, F.-F.; Zhang, S.; Bohrer, A.; Ma, Z.; Turk, J. BBA-Mol. Cell. Biol. L. 2000, 1484, 251-266. McGarry, J. D. Diabetes 2002, 51, 7-18. Stein, D. T.; Stevenson, B. E.; Chester, M. W.; Basit, M.; Daniels, M. B.; Turley, S. D.; McGarry, J. D. J. Clin. Invest. 1997, 100, 398-403.

87

Chapter 4 PILOT STUDY OF WHOLE SEDIMENT PYROLYSIS COMPREHENSIVE TWO-DIMENSIONAL GAS CHROMATOGRAPHY TIME-OF-FLIGHT MASS SPECTROMETERY (GC GC-TOFMS) ON A MEDITERRANEAN SAPROPEL SEQUENCE Introduction A common feature within post-Messinian (5 million years and younger) Mediterranean Sea sediment is dark colored, organic rich layers, known as sapropels, intercalated within surrounding light colored, organic poor layers (marls). Concentrations of total organic carbon (TOC) that can exceed 30 wt% in the sapropels imply both elevated production and enhanced preservation of marine organic matter (OM), neither of which occurs under modern conditions in this oligotrophic sea.1, 2 The exact mechanism for sapropel formation, however, remains under debate and of interest. The periods of time when sapropels were formed correspond to precessional minima (times when the northern hemisphere was tilted towards the sun during the summer, opposite to the current tilt). During the most recent precessional minimum (9-6 thousand years ago), the Mediterranean region was wetter, the Great Plains were drier, and the rainy season of the Asian monsoons was more intense. These and other climate changes might be precursors of what to expect from global warming, which is predicted to cause climate instability and significantly affect how people live today. Paleoclimatic and paleoceanographic information stored in the OM compositions of sapropels and adjacent sediments potentially includes evidence of changes in the origin

88

of the organic matter, its mode and rate of marine productivity, and degree of preservation. As examples, organic geochemical studies have employed Rock-Eval pyrolysis to determine that the bulk of the OM in the sapropels is algal in origin.1, 3, 4 15N values and C/N ratios have been used to establish that microbial nitrogen fixation was important to increased production of marine OM in the sapropels5-7, and to conclude that nitrogen-rich components of OM were selectively recycled in low-oxygen conditions in the water column.1,
3, 6

Recognizing that the extractable molecular components rarely

represent more than a few percent of the total OM, a previous study utilized Curie-point pyrolysis to liberate molecular components from kerogen isolated from three middle Pliocene sapropels and to verify that enhanced preservation of marine OM was indeed important to their formation.4 Gas chromatography is a common tool for the analysis of sapropel extracts. However, the separation of most components is often far from satisfactory as several components tend to coelute and cause large humps or un-unresolved complex mixtures (UCM). The components within the UCMs can only be generalized into a class of molecule but not identified. Comprehensive two-dimensional gas chromatographic (GC GC) analysis is able to better separate and identify the OM in geological samples. This improvement has been demonstrated previously with the analysis of black shale, oil spill environments, diesel fuel, and petroleum.8-15 In Figure 4.1, a direct comparison between a one-dimensional analysis yielding two distinct UCMs and a GC GC analysis of a black shale sample is shown.15 Unlike in the GC analysis where only n-alkanes were identified, in the GC GC chromatogram n-alkanes, mono-, bi-, tri-, tetra- steranes,

89

pentacyclic hopanes, irregular isoprenoids, monomethylalkanes, and monoethylalkanes were identified.

Figure 4.1(A) GCMS TIC chromatogram of black shale containing two UCMs. (B) GC GCMS total ion chromatogram of the same sample with the labeled n-alkanes (black circles), mono-, bi-, tri-, tetra- (steranes), and pentacyclic (hopanes). Dotted line indicates irregular isoprenoids. Boxes encapsulate monomethylalkanes. Open ovals enclose monoethylalkanes. (Used with permission)15

As mentioned previously, extractable molecular components rarely represent more than a few percent of the total OM. Conversely, in pyrolysis, organic material can be analyzed regardless of solubility. Therefore, the use of pyrolysis, instead of liquid extractions, increases the amount of organic matter available for analysis, as demonstrated in Figure 4.216 which compares the GC GC analysis of a limestone

90

Figure 4.2 GC GC chromatogram of an EPA method limestone extraction (top) and pyrolysis GC GC chromatrogram of an identical limestone samples (bottom). In the extract chromatgoram three peaks are present, two are solvent and the third is unidentified. More than 20 peaks were detected in the pyrolysis chromatogram. (Chromatograms courtesy of Dr. Megan McGuigan)

solvent extract with the pyrolysis of limestone. One previous geochemical application of whole sediment Py-GC GC has been done.17 In that study, microwave Py-GC GC with a flame ionization detector was used to study a petroleum source rock. Based on the success of this previous work17, a similar analytical approach is used to identify differences between sapropels and neighboring OM-poor strata because even samples with little OM would contain adequate analyte material if the whole sediment were analyzed. The increased detectability and peak capacity of GC GC combined with

91

pyrolysis was expected to lead to detection of novel organic matter biomarker molecules helping to further elucidate the understanding of sapropel formation. Experimental Samples A sapropel sequence from the Tyrrhenian Basin of the Mediterranean Sea, see Figure 4.3 for location map, was used. The Late Pliocene to Holocene sedimentary record of this semi-isolated sea contains multiple layers of OM-rich sapropels that contrast markedly against the OM-poor homogeneous marls that comprise most of the sediments.18,
19

We obtained samples from Ocean Drilling Program (ODP) Site 974,

which is located at a water depth of 3453 m near the center of the Tyrrhenian Basin. Sediments at this location consist of bioturbated pelagic to hemipelagic nannofossil-rich clays and nannofossil oozes intercaleted sapropel layers.20 The evidence for bioturbation implies that the bottom waters of this basin contained adequate dissolved oxygen to support infauna for at least some parts of its depositional history.

92

Figure 4.3 Location of ODP Site 974 in the Tyrrhenian Basin of the Mediterranean Sea. Depth contours are in meters below sea level.

Our samples were obtained from a sediment sequence shown in Figure 4.4 and includes the sapropel layer that was deposited during insolation cycle 94, which has a mid-point that corresponds to 997 ka.21 This sequence was previously sampled at 1 cm intervals for a study of its bulk OM elemental and isotopic properties.6 A subset of the samples from this earlier study was selected for our pyrolysis studies. The six samples originate from an 18 cm interval that includes the TOC-poor background marl above and below the sapropel and different positions within the TOC-rich layer (Table 4.1). The elevated TOC mass accumulation rates and less negative 13C values of the sapropel samples indicate that they record an interval of elevated marine productivity in the Tyrrhenian Sea. The dramatically low 15N values suggest the existence of nitrogen fixation during deposition, and high C/N values likely represent preferential degradation

93

Figure 4.4 Photo of core used for pyrolysis GC GC analysis. The segment used in this work is highlighted in red. (Photo obtained from the Ocean Drilling Program Data Librarian and used with permission)

of nitrogen-rich forms of OM relative to nitrogen-poor forms. All of these bulk OM parameters are common features of Mediterranean sapropels.1, 3, 5, 6

94

Table 4.1 Samples of ODP Site 974 (Tyrrhenian Basin) insolation cycle 94 sapropel sequence used for pyrolysis GCGC-ToFMS analyses.

Hole-Core-Section, Interval

Depth (mbsf)

Depth (mcd) 56.23 56.40 56.45 56.47 56.51 56.53

Age (ka) 991.5 995.4 996.5 997.0 997.7 998.0

TOC (%) 0.1 0.16 2.51 3.85 2.33 0.76

CorgMAR (mg/cm /ky) 6.4 10.2 160.4 326.5 197.6 64.5

2

d13Corg ( /oo) -24.9 -25.1 -23.0 -22.2 -23.2 -23.4

o

d15Ntot ( /oo) 5.4 3.8 -0.5 -3.2 -0.5 1.9

o

C/N (atomic) 7.4 8.9 17.9 18.6 21.7 18.4

n-alkane ACL9-21 15.5 15.2 15.5 15.7 15.7 12.4

alk-1-ene ACL9-20 14.0* 16.1** 15.3 15.2 14.4 15.4

MK ACL9-19 16.7 15.6 15.9 16.3 15.6 15.5

974C-6H-6, 95-97 cm 974C-6H-6, 112-113 cm 974C-6H-6, 117-118 cm 974C-6H-6, 119-120 cm 974C-6H-6, 123-124 cm 974C-6H-6, 125-126 cm

50.75 50.92 50.97 50.99 51.03 51.05

Bulk sediment elemental and isotopic data from Meyers and Bernasconi (2005) ACL = S(iXi)/S(Xi) where X = peak area and i = range *ACL10-20 **ACL11-20

95

ACL15-17

Pyrolysis comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry (Pyr-GC GC-ToFMS) GC GC-ToFMS studies were performed on whole-sediment samples with the Leco Pegasus III with 4D upgrade and a pyrolysis inlet (Pyroprobe 2000, CDS Analytical, Oxford, PA). Each sample (ca. 2 mg) was placed in a quartz tube between plugs of quartz wool. The tube was inserted into a coil probe and pyrolyzed at 600 C (15 s) under a constant flow of H2. This pyrolysis temperature was determined to be optimal for pyrolysis of our whole sediment samples using the empirical procedure of McGuigan et al.22 Briefly, GC GC chromatograms were obtained for fresh portions of the 119-120 cm interval sample at pyrolysis temperatures of 200, 400, 600, 800 and 1000 C. These chromatograms were then compared to those obtained from one 2 mg sample of the same interval that was pyrolyzed and analyzed at each temperature in increasing order (200 to 1000 C). Between analyses, the quartz tube was removed immediately after pyrolysis and stored in a sealed vial until the previous chromatogram was complete, and the interface was purged to avoid carry-over. The 600 C pyrolysis temperature that we empirically determined to be optimal for our whole-sediment analyses is very close to the Curie point pyrolysis temperature (610 C) used in a previous study4 of sapropel kerogens but somewhat below the microwave pyrolysis temperature (650 C) used in a whole-sample study of a petroleum source rock.17 Pyrolysis products were transferred into the GC injection port through a heated interface. It is important to note that the valve releasing the pyrolysis products into the GC inlet was manually controlled, which introduced slight differences in GC retention times from sample to sample. An RTX-1 column (Restek Corporation, Bellefonte,PA) 96

was used in the first dimension (30 m, 0.25 mm i.d., 0.1 m film) and an RTX-Wax column (Restek Corporation, Bellefonte, PA) in the second dimension (2 m, 0.10 mm i.d., 0.1 m film). The temperature of the first dimension column oven was initially 40 C (5 min hold) and was increased by 4 C min-1 to 200 C (20 min hold). The second dimension column was housed in an oven that had a 5 C offset from the main oven, and the modulator had a 30 C offset. The ToFMS was operated at 70 eV with a mass range of m/z 35-350 with spectra collected at 100 Hz. The modulation period used was 4 seconds. Instrument control and data processing were performed using ChromaTOF software version 4.22 (Leco Corporation). Data were processed with a signal/noise (S/N) ratio of 75 and 25, a baseline offset of 0.5, and automatic smoothing. Peak identification was done by automated comparison with the National Institute of Standards and Technology (NIST) 02 spectral library, augmented with literature references.4, 8, 9, 23 Results and discussion Column Selection and Chromatographic Conditions An RTX-Wax column was chosen for the second dimension because it is orthogonal to the non-polar RTX-1 used in the primary dimension and was readily available at the time of the experiment. As can be seen in Figure 4.5A, temperature limits of the polar column caused some chromatographic issues. The upper limit of 250 C (~C22) for the RTX-Wax column forced an isothermal hold at the end of the chromatographic run in order to elute as many higher boiling components as possible. This constraint causes an increase in second dimension retention times due to the high capacity factor (k) of these components. In addition, this limitation could possibly have led to analytes not completing elution during the modulation cycle in which they were 97

injected and eluting instead in the next modulation cycle, also known as chromatographic wrap-around. However, as the components are resolved in our chromatograms, this behavior was not problematic and simply needs to be recognized as a potential issue in future studies. In fact, one could argue that allowing for some wrap-around is an effective use of the chromatographic space rather than leaving two seconds empty in the second dimension. Peak tailing in the first dimension is also very apparent in peaks such as benzene and toluene (Figure 4.5A). This behavior is likely due to large amounts of these components being present in some samples and thus overloading the first dimension and not experiencing an isothermal separation in the second dimension. As the temperature of the second dimension changes, the temperature that the tail of the peak is experiencing changes, thus decreasing the k value and causing the peak to tail downwards rather than elute as a single wide peak. Results of whole sediment pyrolysis comprehensive two-dimensional gas

chromatography time-of-flight mass spectrometry (Py-GC GC-ToFMS) analyses Representative GC GC total ion chromatograms of an OM-rich (TOC 3.85 %) interval (119-120 cm) and an OM-poor (TOC 0.1 %) interval (95-97 cm) are shown in Figures 4.5A and 4.6A, respectively. Differences exist between the pyrolysates of the bulk OM present in the sapropel layer and in the sediment 22 cm above it and are most obvious when comparing the expanded specific mass chromatograms shown in Figures 4.5B through 4.5I and 4.6B through 4.6I. We discuss some of the details of the results of our whole sediment pyrolytic analyses below.

98

Figure 4.5 GC GC total ion chromatogram (TIC) of sapropel interval 119-120 cm (A). Expanded regions of interval 119-120 cm: benzene region (B), representative alkene region (C), furan region (D), thiophene region (E), pyrroles (F), naphthalenes (G), phenol region (H), and methyl ketones (MKs) (I). All traces are TIC unless otherwise indicated.

99

Figure 4.6 GC GC total ion chromatogram (TIC) of non-sapropel interval 112-113 cm. Expanded regions of interval 112-113 cm: benzene region (B), representative alkene region (C), furan region (D), thiophene region (E), pyrroles (F), naphthalenes (G), phenol region (H), and methyl ketones (MKs) (I). All traces are TIC unless otherwise indicated.

100

Alkanes and alkenes Straight chain alkanes were identified in all the two-dimensional chromatograms, as represented by Figures 4.5A and 4.6A. They range from C8 to C21, similar to the range observed previously4 in a pyrolytic study of three Pliocene sapropels. The average chain length (ACL)9-214 of our Tyrrhenian Basin sapropel pyrolysates are 15.5, 15.7, and 15.7 for intervals 117-118 cm, 119-120 cm, and 123-124 cm, respectively (Table 4.1). The ACL9-21 values of the marl samples, which are 15.5, 15.2, and 12.4 for intervals 95-97 cm, 112-113 cm, and 125-126 cm, see Table 4.1, are similar to those found in the sapropels, although slightly lower on average. In contrast to our results, Menzel et al. observed that sapropel ACL values were larger than those of TOC-poor marls in the pyrolysates of Pliocene sequences from the eastern Mediterranean.4 We believe that this difference in results is likely to represent real differences between the sapropel depositional conditions in the Pleistocene Tyrrhenian Basin and the Pliocene eastern Mediterranean Sea. Although we cannot give a specific explanation of the nature of the depositional differences, we speculate that slower ventilation of the deep Mediterranean during the Pliocene than in the middle Pleistocene24 may have limited oxidation of marine OM in the Pliocene sapropels and contributed to the contrasts in the ACL patterns via somewhat better preservation of their OM. Eluting just prior to the n-alkanes, n-alk-1-monoenes with the same chain lengths occur, as shown in Figures 4.5C and 4.6C, while n-alk-monoenes with a non-terminal double bond elute after the corresponding n-alkanes. ACL9-20 values of the alk-1-enes are 14.0, 16.1, 15.3, 15.2, 14.4 and 15.4 for intervals 95-97 cm, 112-113 cm, 119-120 cm, 123-124 cm, and 125-126 cm, respectively. It should be noted however that the ACL 101

values for Intervals 95-97 cm and 112-113 cm are over the carbon ranges of 10-20 and 11-20, respectively, due to a lack of detectable C9 and C10 alk-1-enes in these samples. In addition, C13 and C15 alk-1-enes were identified only in interval 117-118 cm after the sample was processed at a signal/noise ratio of 25. The short chain n-alkanes and n-alk-1-enes in the pyrolysates have been postulated to derive from the algaenan in the cell walls of several classes of the marine algae that contributed OM to Pliocene sapropels.4 The greater abundance of even-chain n-alkanes and n-alkenes than their odd-chain analogs in both the sapropels and the marls (Figs. 4.5 and 4.6) is similar to previous results4 and may well derive from pyrolysis of algal organic matter. Further, the relatively small ACL values in pyrolysates of the Pleistocene Site 974 sequence may indeed suggest that both the sapropels and the marls contain large proportions of algal-derived OM, an interpretation that is supported by greater contributions of n-alkanes and n-alk-1-enes in the sapropel samples relative to the marls (Table 4.2). However, our use of the RTX-Wax column limited our analysis to n-alkanes < C22, so we were not able to assess contributions of land-plant waxes to this sapropel sequence. Two isomers of alk-2-enes were detected in most of the samples (Table 4.3). One isomer is clear in both Figures 4.5C and 4.6C, eluting just after the alkane peak in the first dimension. The second one, however, is more difficult to identify because it essentially coelutes with alkanes at these retention times. However, when looking specifically at the C10 through C13 regions in Figure 4.5, one can easily identify four peaks, three across the top (alk-1-ene, alk-2-ene, alk-2-ene) and one below the center

102

Table 4.2 Alkanes (x) and branched alkanes () identified in the respective intervals. ND = not detected. At least one isomer of branch ed alkanes was identified where indicated by (). 95-97 cm TOC 0.10 % Alkane C8 C9 C10 C11 C12 C13 C14 C15 C16 X X X X X X X X X X X X X X Branched ND ND ND ND ND ND 112-113 cm TOC 0.16 % Alkane X X X X X X X X X X X X X X Branched ND ND ND ND ND ND ND ND ND ND ND ND ND 117-118 cm TOC 2.51 % Alkane X X X X X X X X X X X X X ND Branched ND ND ND 119-120 cm TOC 3.85 % Alkane X X X X X X X X X X X X X X Branched ND ND ND 123-124 cm TOC 2.33 % Alkane X X X X X X X X X X X X X ND Branched ND ND ND ND ND ND 125-126 cm TOC 0.76 % Alkane X X X X X X X X X X X X X ND Branched ND ND ND ND ND ND ND ND

103

C17 C18 C19 C20 C21

Table 4.3 Alk-1-enes and alk-2-enes identified in respective intervals. X indicates a visible peak but the absence of a software peak marker. ND = not detected 95-97 cm TOC 0.10 % Alk-1-ene C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 ND X X X X X X X X X X Alk-2-ene ND ND X X X X X X X X ND 112-113 cm TOC 0.16 % Alk-1-ene ND ND X X X X X X X X X Alk-2-ene ND X X X X XX X X X X ND 117-118 cm TOC 2.51 % Alk-1-ene X X X X X* X X* X X X X Alk-2-ene ND X XX XX XX xx XX XX XX XX X ND 119-120 cm TOC 3.85 % Alk-1-ene X X X X X X X X X X X X Alk-2-ene X XX XX XX XX XX XX XX XX XX X ND 123-124 cm TOC 2.33 % Alk-1-ene X X X X X X X X X X X X Alk-2-ene X XX XX XX XX XX XX XX XX XX ND ND 125-126 cm TOC 0.76 % Alk-1-ene X X X X X X X X X X X X Alk-2-ene ND XX XX XX XX XX XX XX XX XX X ND

104

C19

X ND X ND X C20 * Needed to be reprocessed with a S/N = 25 in order to get peak markers

peak, the alkane. In a number of instances, as indicated by the data in Table 4.3, the peaks were visible but were not identified by the software. Furans and thiophenes Sapropel intervals 117-118 cm, 119-120 cm, and 123-124 cm yielded 2,5 or 2,4-dimethyl (not in 123-124 cm), 2-ethyl-5-methyl, 2,3,5-trimethyl, and

4-methyl-2-propyl furans (Table 4.4). Figure 4.5E also has two unknown peaks (U3 and U4) as well as a peak identified by the software as 1,2,3-cyclopentene. Although it is not clear what U4 is, U3 is very likely an isomer of dimethlyfuran that was not identified as a peak by the software, possibly because of a low S/N ratio. The 1,2,3-cyclopentene peak is likely a rearrangement product as no real precursors for this peak are known. It has been proposed that furans are produced by pyrolysis of carbohydrates and that their presence indicates better preservation of labile OM in sapropels than in TOC-poor marls.25 The absence of a large variety of these compounds in our marl sequence (Figure 4.6D and Table 4.4) is consistent with previous interpretation.25 Methyl, ethyl, dimethyl, butyl, and propyl thiophenes were identified in all of the pyrolysis samples, with the exception of intervals 95-97 cm and 125-126 cm (Table 4.4). These compounds appear as the m/z 97 traces in Figures 4.5E and 4.6E and, like the furans, are thought to be carbohydrate pyrolysis products.25 However, they are formed by incorporation of S2- into their carbohydrate precursors and therefore reflect existence of euxinic conditions at the sea floor or within the upper sediment during sapropel deposition.26 Such conditions are likely to improve organic matter preservation by a combination of stabilization of reactive forms of organic matter via sulfurization and absence of molecular oxygen that would otherwise allow infaunal grazing on the OM. 105

Table 4.4 Furans, thipohenes, and pyrroles, identified in respective intervals. ND = not detected, X = detected. TOC is in percent. 95-97 cm TOC 0.10 Furans 2,5 or 2,4-dimethyl 2-ethyl-5-methyl 2,3,5-trimethyl 4-methyl-2-propyl Thiophenes 2-methyl 3-methyl 2-ethyl 3-ethyl 2,5-dimethyl 2,4-dimethyl 2,3-dimethyl butyl propyl Pyrroles 1-methyl-1H 1-ethyl-1H 2,5-dimethyl 2,3-dimethyl 2-ethyl-4-methyl 2,3,4-trimethyl 2,3,5-trimethyl X ND ND ND ND ND ND ND ND ND ND ND ND ND X ND X ND X X X X X X X X X X ND X X X X X X ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND X X X X X X X X ND X X X ND ND ND ND 112-113 cm TOC 0.16 117-118 cm TOC 2.51 119-120 cm TOC 3.85 123-124 cm TOC 2.33 125-126 cm TOC 0.76

The existence of both furans and thiophenes in the pyrolysis products is evidence of improved organic matter preservation. Pyrroles A suite of pyrroles, consisting of 1-methyl, 2 and 3-ethyl, 2,5-dimethyl, 2,4-dimethyl, 2,3-dimethyl, 2,3,4-trimethyl, and 2,3,5-trimethyl was found in the three sapropel samples but not in the marl samples (see Figures 4.5F and 4.6F and Table 4.4). Previous results similarly report pyrroles in the pyrolysates of Pliocene sapropel samples

106

from three locations in the eastern Mediterranean and not in the surrounding marl.4 It has been proposed that these alkyl pyrroles are likely derived from tetrapyrrole pigments such as chlorophyll.25 The relatively labile nature of tetrapyrrole pigments argues against their delivery from continental sources.4 Instead, the common presence of these chlorophyll derivatives in both the Pleistocene (1 Mya) and Pliocene (3 Mya) sapropels from four different locations suggests that heightened production and improved preservation of marine OM were typically linked during times of sapropel deposition. Alkyl aromatics Alkyl aromatics in pyrolysates can have many precursors from the mix of material that can constitute kerogen.27 We found benzene, toluene, indane, methyl and dimethylindanes, naphthalenes, and methylated and alkylated benzenes in all of the pyrolysates of the Hole 974C sequence (Tables 4.5 & 4.6). Toluene has been reported as a pyrolysis product of humic acids.26 It is likely present in both the sapropel and marl samples because humic acids from both terrigenous and marine sources produce toluene when pyrolyzed.26

107

Table 4.5 Naphthalenes and phenols identified in respective intervals. ND = not detected, X = detected 95-97 cm TOC 0.10 % Naphthalenes naphthalene 2-methyl 1-methyl 2,6-dimethyl 2,7-dimethyl 1,3-dimethyl 1,7-dimethyl 1,6-dimethyl 1,5-dimethyl 1,2-dimethyl Phenols phenol methyl ethyl ND ND ND ND ND ND X X X X X ND X X X ND ND ND X X X X X X X X X ND X X X X X X X X X ND X X X X X X X X X ND X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 112-113 cm TOC 0.16 % 117-118 cm TOC 2.51 % 119-120 cm TOC 3.85 % 123-124 cm TOC 2.33 % 125-126 cm TOC 0.76 %

108

Table 4.6 Benzene and indane isomers identified in respective intervals. ND = not detected, X = detected. TOC is in percent. 95-97 cm TOC 0.10 Benzene Toluene Ethyl benzene m-xylene + p-xylene o-xylene Sub. Benzenes isopropyl propyl 1-ethyl-3-methyl 1-ethyl-4-methyl 1,3,5-trimethyl 1-ethyl-2-methyl 1,2,4-trimethyl 1,2,3-trimethyl butyl 1-methyl-2-propyl 1-ethyl-2,4-dimethyl2-ethyl-1,4-dimethyl1,2,4,5-tetramethyl 1,2,3,4-tetramethyl pentyl 1-methyl-4-(2-methylpropyl)hexylbenzene Indanes Indane methyl dimtheyl dimethyl X X X X X ND X ND X X X X X X X X X X X ND X ND ND ND ND X X ND X ND X ND X X X X ND ND X X X ND X ND X ND X X X X X ND ND X ND X X X ND X X X X X X X X X X ND X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X ND X X ND X X X X X X X ND X X X X X X X X X X 112-113 cm TOC 0.16 X X X X X 117-118 cm TOC 2.51 X X X X X 119-120 cm TOC 3.85 X X X X X 123-124 cm TOC 2.33 X X X X X 125-126 cm TOC 0.76 X X X X X

Methyl and dimethyl naphthalenes were also found in the six pyrolysates (Table 4.5). A typical distribution is highlighted at m/z 128 +141 in Figures 4.5G and 4.6G. The presence of naphthalenes may be evidence of Type II kerogen and is likely formed in marine settings with a major source consisting of autochthonous OM from phytoplankton with smaller contributions of allochthonous higher plant matter.28 Indane, 109

methylindane, and dimethylindane were identified in both the sapropel and the marl samples, as shown in Figures 4.5B and 4.6B. Ethylbenzene, meta, para and ortho-xylene were also identified in all of the samples, see Figures 4.5B and 4.6B. Interestingly, as can be seen in Figure 4.5B and slightly less so in 4.6B, in all instances there were additional peaks immediately following the main xylene peaks in the first dimension, U1 and U2. Although these peaks appear to be quite intense, particularly in Figure 4.5B, they either do not contain peak markers or are identified as xylenes. It is thought that peaks U1 and U2 are merely peak tails of the xylene components, however, it is not understood why a gap exists between the peak and the tail in a number of chromatograms. The alkylated benzene region is presented in Figures 4.5B and 4.6B. Trimethyl, tetramethyl, pentamethyl, and hexamethyl benzenes are highlighted in both figures. The trimethyl and tetramethyl benzenes are more prevalent than the pentyl and hexyl methyl benzenes as indicated by the trimethyl and tetramethyl benzenes having an average peak area an order of magnitude greater than that of the pentyl and hexyl benzenes (data not shown). The dominance of trimethyl and tetramethyl benzenes in sapropel layers agrees with previous reports.4, 27 The pattern is consistent with the Rock-Eval identification of Type II or Type II S kerogen in Pleistocene sapropels at Site 974 reported previously.3 Also, as shown previously,27 the absolute amount of alkyl benzenes obtained as flash pyrolysis products is greater from Type II and Type II S kerogen than from Type I and Type III kerogen. Moreover, the prevalence of tetramethyl benzenes (specifically 1, 2, 3, 4- and 1, 2, 3, 5-tetramethyl benzene) is a pyrolysis biomarker for kerogen-bound carotenoids, particularly isorenieratene, that are derived from green sulfur bacteria, which 110

are nitrogen-fixing obligate anaerobes.27 1,2,3,4-tetramethyl benzene is found in the three sapropel samples we pyrolyzed and in the closely underlying marl sample, but not in the overlying marl samples (Table 4.6). The presence of this postulated derivative of isorenieratene indicates that the cycle 94 sapropel layer was created under euxinic conditions in the photic zone of the Tyrrhenian Sea, as concluded by multiple investigators for sapropel layers elsewhere in the Mediterranean basin. Phenols Phenol, methyl phenol, and ethyl phenol were identified in the sapropel intervals 117-118 cm, 119-120 cm and 123-124 cm, see Figure 4.5H. In Figure 4.6H (non-sapropel interval 95-97 cm), it can be seen that the phenols are missing, which was also the case for the intervals 112-113 cm and 125-126 cm. The peaks that are visible in Figure 4.6H are the xylenes and the associated tails discussed previously and labeled in Figure 4.6B. Methyl Ketones Methyl ketones C9 through C19 were identified in all of the pyrolysates (Table 4.7) with the exception of interval 95-97 cm. It should be noted, however, that although the peaks are identified in interval 112-113 cm the S/N was notably lower for this sample compared to the others (S/N ~40 compared to S/N >75). Interval 112-113 cm had to be reprocessed in ChromaTOF with S/N threshold of 25 in order to obtain identification of the methyl ketones. Methyl ketones have been reported in a number of marine sedimentary settings
29, 30 4,

yet their origin remains enigmatic. Because their molecular distributions are usually

dominated by odd-chain-length components, they have been postulated to be the products

111

Table 4.7 Methyl ketones identified in respective sapropel samples. ND = not detected, X indicates that a peak was identified by the software 95-97 cm TOC 0.10 % C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 ND ND ND ND ND ND x x x ND ND 112-113 cm TOC 0.16 % x x x x x x x x x x x 117-118 cm TOC 2.51 % x x x x x x x x x x x 119-120 cm TOC 3.85 % x x x x x x x x x x x 123-124 cm TOC 2.33 % x x x x x x x x x x x 125-126 cm TOC 0.76 % x x x x x x x x x x x

either of microbial oxidation of odd-chain-length n-alkanes29 or of microbial decarboxylation of even-chain-length n-alkanoic acids.30 Because of their common presence in the pyrolysates of the Site 974 samples, we conclude that these molecular fragments likely represent the straight carbon chains of biolipids that have survived since these sediments were deposited 1 Mya. The ACL9-19 values of both the sapropels (15.9 mean) are slightly larger than those of the marls (15.5), which neither value is significantly different from the ACL values of the n-alkanes or the n-alkenes (Table 4.1). Potential Benefits of Py-GC GC-ToFMS The often mentioned benefits of GC GC are increased peak capacity, increased detectability, and structured chromatograms. The benefit of peak capacity is the most obvious in extremely complex samples such as diesel oil11 where thousands of peaks can be identified and most would not be resolved without the extra chromatographic space provided. This extra space is also beneficial in the less complex samples described here, however. For example, in a number of cases the alk-2-ene peak that would have co-eluted 112

with the corresponding alkane peaks is easily identified as a separate analyte even without the use of mass spectral deconvolution. In addition, analytes eluting later in the second dimension, such as the naphthalenes, do not co-elute with the substituted benzenes or the alkane/alkenes with similar first dimension retention times. Structured chromatograms were beneficial in the identification of peaks in the sapropel samples. The structure obtained from GC GC can be seen clearly in Figures 4.5 and 4.6 where similar compounds such as the C3 through C6-benzenes elute as a group along the x-axis and the isomers of each carbon number elute as separate groups along the y-axis. The increased detectability is demonstrated by the methyl ketones and the alkenes, where peaks were only identified by the software after reprocessing with a S/N threshold of 25. The reprocessing capability indicates that these peaks are only present in trace amounts and may have gone undetected in one-dimensional gas chromatographic analysis. It is estimated that a 30-fold increase in sensitivity can be expected in GC GC31 and with further optimization it is likely that this technique could be used to successfully identify previously undetected analytes in sapropels and other geological samples. A benefit of pyrolysis is that it releases organic fragments that have been incorporated into macromolecular organic matter and possibly also the mineral matrix of sediments. This benefit is, of course, tempered by the likelihood that the molecular constitution of the original material has been altered first by diagenesis and again by pyrolysis. Nonetheless, useful information about the origin and alteration of the organic constituents can be inferred from the pyrolysis products.

113

Conclusions Our evaluation of GC GC-ToFMS analysis of the molecular fragments released by pyrolysis of whole sediment samples of a middle Pleistocene sapropel sequence shows that this analytical approach can yield important information about the delivery and preservation of OM during sapropel deposition. In agreement with previous work and indicative of a wetter climate and a stratified and more productive water column, we found evidence of elevated marine productivity during the deposition of the sapropel layer as suggested by the presence of alkyl pyrroles in the sapropel pyrolysates but not in those of the marls. These compounds are considered the pyrolysis products of tetrapyrrole pigments like chlorophyll, which are unlikely to survive transport from land sources and are thus produced locally in the surface ocean. Improved preservation of marine OM under conditions of diminished oceanic circulation is indicated by higher abundances of shorter chain aliphatic hydrocarbons, pyrroles, furans, and alkyl aromatics in the sapropel pyrolysates than in the marls. The presence of phenols possibly derived from lignin in the sapropel pyrolysates but not in those of the marls may indicate overall improved preservation of OM in general. The more common presence of thiophenes in the pyrolysates of the sapropel samples than in those of the marls records the existence of euxinic conditions at or near the sea floor during sapropel deposition. Tetramethyl benzenes, likely derived from isorenieratene, are present in the sapropel samples but not in the marls, recording euxinic conditions within the photic zone during the period of sapropel formation. Although a broad spectrum of products was released by pyrolysis of the OM in the whole sediment samples and a large number of the components were successfully 114

separated and identified using GC GC, the original goal of detecting and identifying novel organic matter biomarkers was not fulfilled. However, paleoclimatic and paleoceanographic information was obtained for this sapropel sequence without the need for time and solvent intensive liquid extractions. This work demonstrates the first example of whole sediment pyrolysis GC GC analysis of a sapropel sequence as well as the potential for continued use of this technique when studying other whole sediment samples. Based on our evaluation of the GC GC-ToFMS pyrolysis analysis of the whole sediment samples, we suggest that future similar analyses include several refinements to produce the best results. The optimal pyrolysis conditions should be identified using the experimental protocol described previously.22 We established that 15 s at 600 C were the optimal conditions for our sapropel sequence samples, but a different time at a higher or lower temperature might be more suitable for other types of whole sediment samples. Operation of the transfer valve from the pyrolysis inlet into the first dimension GC column could be automated to obtain identical transfer times. We operated this valve manually, which introduced slight differences in GC retention times and complicated identifications of pyrolysates. Stationary phases of both the first dimension and second dimension columns should be selected to optimize analysis of the sample type and the analytes of interest. We were obligated to use an RTX-Wax for our second dimension column because of shared uses of the Pegasus system. This phase limited oven temperatures to 250 C and hence constrained our analysis to pyrolysis fragments no larger than 22 carbon atoms. A higher temperature capability for the second column would have been more suitable for our sapropel sequence. 115

References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Bouloubassi, I.; Rullkotter, J.; Meyers, P. A. Mar. Geol. 1999, 153, 177-197. Rinna, J.; Warning, B.; Meyers, P. A.; Brumsack, H. J.; Rullkotter, J. Geochim. Cosmochim. Acta 2002, 66, 1969-1986. Meyers, P. A.; Doose, H. Proceedings, Ocean Drilling Program, Scientific Results 1999, 161, 383-390. Menzel, D.; van Bergen, P. F.; Veld, H.; Brinkhuis, H.; Damste, J. S. S. Org. Geochem. 2005, 36, 1037-1053. Milder, J. C.; Montoya, J. P.; Altabet, M. A. Proceedings, Ocean Drilling Program, Scientific Results 1999, 161, 401-411. Meyers, P. A.; Bernasconi, S. M. Mar. Geol. 2005, 220, 41-58. Higgins, M. B.; Robinson, R. S.; Carter, S. J.; Pearson, A. Earth. Planet. Sci. Lett., 290, 102-107. Frysinger, G. S.; Gaines, R. B. J. Sep. Sci. 2001, 24, 87-96. Gaines, R. B.; Frysinger, G. S.; Hendrick-Smith, M. S.; Stuart, J. D. Environ. Sci. Technol. 1999, 33, 2106-2112. Nelson, R. K.; Kile, B. M.; Plata, D. L.; Sylva, S. P.; Xu, L.; Reddy, C. M.; Gaines, R. B.; Frysinger, G. S.; Reichenbach, S. E. Environmental Forensics 2006, 7, 33-44. Wang, F. C. Y.; Robbins, W. K.; Greaney, M. A. J. Sep. Sci. 2004, 27, 468-472. vila, B. M. F.; Aguiar, A.; Gomes, A. O.; Azevedo, D. A. Org. Geochem., 41, 863-866. Pomerantz, A. E.; Ventura, G. T.; McKenna, A. M.; Canas, J. A.; Auman, J.; Koerner, K.; Curry, D.; Nelson, R. K.; Reddy, C. M.; Rodgers, R. P.; Marshall, A. G.; Peters, K. E.; Mullins, O. C. Org. Geochem., 41, 812-821. Ventura, G. T.; Raghuraman, B.; Nelson, R. K.; Mullins, O. C.; Reddy, C. M. Org. Geochem., 41, 1026-1035. Ventura, G. T.; Kenig, F.; Reddy, C. M.; Frysinger, G. S.; Nelson, R. K.; Van Mooy, B.; Gaines, R. B. Org. Geochem. 2008, 39, 846-867. McGuigan, M. Dissertatioin, University of Michigan, Ann Arbor, MI, 2005. Wang, F. C. Y.; Walters, C. C. Anal. Chem. 2007, 79, 5642-5650. Cramp, A.; O'Sullivan, G. Mar. Geol. 1999, 153, 11-28. Emeis, K. C.; Sakamoto, T.; Wehausen, R.; Brumsack, H. J. Palaeogeography Palaeoclimatology Palaeoecology 2000, 158, 371-395. Comas, M. C.; Zahn, R.; Klaus, A.; Aubourg, C.; Belanger, P. E.; Bernasconi, S. M.; Cornell, W.; de Kaenel, E. P.; de Larouziere, F. D.; Doglioni, C.; Doose, H.; Fukusawa, H.; Hobart, M.; Iaccarino, S. M.; Ippach, P.; Marsaglia, K.; Meyers, P.; Murat, A.; O'Sullivan, G. M.; Platt, J. P.; Prasad, M.; Siesser, W. G.; Skilbeck, C. G.; Soto, J. I.; Tandon, K.; Torii, M.; Tribble, J. S.; Wilkens, R. H.; Riegel, R. N. Annotation: Individual papers are cited separately 1996, 161, 1023. Laskar, J.; Joutel, F.; Boudin, F. Astron. Astrophys. 1993, 270, 522-533. McGuigan, M.; Waite, J. H.; Imanaka, H.; Sacks, R. D. J. Chromatogr. A 2006, 1132, 280-288. Budzinski, H.; Raymond, N.; Nadalig, T.; Gilewicz, M.; Garrigues, P.; Bertrand, J. C.; Caumette, P. Org. Geochem. 1998, 28, 337-348. 116

(11) (12) (13)

(14) (15) (16) (17) (18) (19) (20)

(21) (22) (23)

(24) (25) (26) (27) (28) (29) (30)

(31)

Gallego-Torres, D.; Martinez-Ruiz, F.; Meyers, P. A.; Paytan, A.; JimenezEspejo, F. J.; Ortega-Huertas, M. Biogeosciences 2010, In Press. Sinninghe Damste, J. S.; Kok, M. D.; Koster, J.; Schouten, S. Earth. Planet. Sci. Lett. 1998, 164, 7-13. Dick, C.; Ediger, V.; Fabbri, D.; Gaines, A. F.; Love, G. D.; McGinn, A.; McRae, C.; Murray, I. P.; Nicol, B. J.; Snape, C. E. Fuel 2002, 81, 431-448. Hartgers, W. A.; Damste, J. S. S.; Deleeuw, J. W. Geochim. Cosmochim. Acta 1994, 58, 1759-1775. Killops, S.; Killops, V. Introduction to Organic Geochemistry, 2nd ed.; Blackwell Publishing: Malden, MA, 2005. Simoneit, R. T.; Mazurek, M. A.; Brenner, S.; Crisp, P. T.; Kaplan, I. R. DeepSea Research 1979, 26, 879-892. Volkman, J. K.; Farrington, J. W.; Gagosian, R. B.; Wakeham, S. G. Lipid composition of coastal marine sediments from the Peru upwelling region.; Wiley: New York, 1983. Marriott, P.; Shellie, R. Trac-Trends in Analytical Chemistry 2002, 21, 573-583.

117

Chapter 5 SUMMARY AND FUTURE WORK Summary A metabolite profiling and a metabolomics method utilizing comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry

(GC GC-ToFMS) was developed and applied to INS-1 cells, an insulin-secreting cell line commonly used as a model in diabetes studies. Twenty seven target metabolites were identified at low, moderate and high glucose concentrations with an average technical relative standard deviation (RSD) of 18, 17 and 14 % and biological RSDs of 27, 24 and 20 % for 3 mM, 7 mM and 17 mM glucose respectively. Thirty two metabolic pathways were identified in the metabolomic analysis, nine of which are discussed in relation to a previous or a potential relevance to glucose stimulated insulin release, -cell dysfunction or diabetes. This work demonstrates the utility of this technique for both targeted and global metabolic analysis as well as identifies potential target analytes and pathways for future experiments. A GC GC-ToFMS method was also developed for the determination of total fatty acids in cultured mammalian cells. Like the previous metabolomic profiling and metabolomic study, this method was applied to INS-1 cells. In this method, lipids were extracted and transformed to fatty acid methyl esters for analysis. GC GC analysis revealed the presence of 30 identifiable fatty acids in the extract. This result doubles the number of fatty acids previously identified in these cells.1, 2 The method yielded linear 118

calibrations and an average relative standard deviation of 8.4 % for replicate injections of samples and 12.4 % for replicate analysis of different samples. The method was used to demonstrate changes in fatty acid content as a function of glucose concentration on the cells. The results demonstrated the utility of this method for analysis of fatty acids in mammalian cell cultures. Evaluation of the use of GC GC-ToFMS to analyze the pyrolysis products of six whole sediment samples obtained from above, within, and below a 1 million year old organic matter (OM) rich Mediterranean sapropel layer was performed. Analysis of whole sediment samples can avoid analytical bias that might result from isolation of OM components from the sediment matrix. Although the temperature limit of the second GC column used constrained the analyses to pyrolysis products containing <22 carbon atoms, differences were found between the OM-rich sapropel samples and the OM-poor background marls. The presence of alkyl pyrroles, which are probably derived from chlorophyll, in pyrolysates of the sapropels but not in those of the marls indicates higher marine productivity and greater OM preservation accompanied deposition of the sapropels. Detection of tetramethyl benzenes that are considered to be pyrolysates derived from isorenieratene in the sapropel samples is evidence that nitrogen-fixing green sulfur bacteria contributed to the high productivity. Greater abundances of shorter chain aliphatic hydrocarbons, pyrroles, furans, and alkyl aromatics in the pyrolysates of sapropel samples relative to the marls confirm better preservation of marine OM in the sapropels. In addition, the presence of larger amounts of thiophenes in the sapropels than in the marls is consistent with the existence of euxinic conditions during sapropel

119

deposition. The combination of whole-sediment pyrolysis and GC GC-ToFMS is promising, but requires careful selection of pyrolysis temperature and column selection. Future Work Metabolite Profiling and Metabolomics Analysis The results presented in Chapter 2, specifically the metabolomics analysis, could lead to the inception of a number of different experiments to further probe the effect of glucose on the INS-1 cell line. Given that arachidonic acid (AA) and AA metabolism are involved in a number of the pathways discussed, it would be reasonable to make AA and its metabolite analytes of interest. The experiment performed in Chapter 2 could be repeated with AA instead of glucose in order to probe the effects of this analyte on INS-1 cell metabolism. Alternatively, a variety of experiments could be performed to test the theory that AA has a positive effect on insulin secretion in -cells. For example, after treatment with high glucose, some of the cells could be treated with AA. Alternatively, cells could be treated with palmitic acid followed by AA. In both cases, metabolite profiling and metabolomic analysis similar to that performed in Chapter 2 could be used for analysis. INS-1 cells could also be selectively treated with specific COX and LOX inhibitors prior to glucose treatment, and a similar analysis performed. The COX and LOX inhibitors needed could be determined by performing a PCR analysis of the cells to determine which COX and LOX enzymes are expressed in the INS-1 cell line used. It has been shown previously that the enzymes expressed in human and rat islets vary slightly. Therefore, it is reasonable to assume that there is also variation in the enzymes expressed

120

in the INS-1 cell line.3 Finally, through all of the experiments suggested above, prostaglandin E(2) (PGE(2)) could be a target metabolite. It has been shown previously that this analyte is detectable using GC/MS and a derivatization method similar to the one used in Chapter 2.4 Another pathway that would be interesting to investigate further is the butanoate metabolism pathway. Given previous results that dietary butanote improve insulin sensitivity in mice,5 it would be interesting to simultaneously treat INS-1 cells with butanoate and glucose followed by an analysis similar to that described in Chapter 2. If the method developed in Chapter 2 could be adopted for islets studies, as is discussed below, it would also be interesting to perform a study identical to that presented in Chapter 2 using islets from mice that had and had not received dietary butanoate. Propanoate metabolism is also a potential pathway for future interest. INS-1 cells could be incubated in propanoate prior to extraction followed by metabolite profiling and metabolomics analysis. Changes in the current set of target metabolites would be expected given that propanoate can enter the citric acid cycle via succinyl CoA. Additionally, treatment with methyl-succinate, which can cross into the mitochondria to be converted to succinyl CoA, may also be interesting. An interesting follow up experiment for either of these secretagogues would be to block the conversion of succinyl CoA, by knocking down methymalonyl CoA mutase, and repeating the initial experiment.

121

Analysis of Lipids The advantages of utilizing GC GC for lipid analysis can be further exploited for the separation of lipid classes. Lipids can be separated into cholesterol ethers, triglycerides, free fatty acids, and phospholipids through the use of thin-layer chromatography on silica gel plates.2 Although this would require an alternative method of fatty acid methyl ester synthesis, it would also allow for analysis of redistribution between lipid classes. Additionally, the phospholipids can be further separated into phosphatidylcholines, phosphatidylethanolamines, phosphatidylinositol, and

phosphatidylserines2 allowing for even more detailed analysis of the fatty acid redistribution. Quantifaction of FAMEs could be improved by using stable isotope labeling. Similar to previous work,6 FAMEs synthesis, as described in Chapter 3, could be completed on plates of INS-1 cells using both methanol and deuterated methanol. A single plate at a given concentration would be synthesized with the deuterated methanol allowing it to serve as in internal standard for all of the other samples. An aliquot of the deuterated extract could then be combined with an aliquot of an unlabelled extract to create a sample that has an isotope labeled standard for every analyte. Samples could then be analyzed by GC GC as previously described and quantification would be based on peak area ratios. Preliminary work in this area has been completed and a sample

chromatogram is shown in Figure 5.1. The overall chromatography is quite similar to

122

Figure 5.1 Fatty acid methyl esters in INS-1 cells combined with an isotopically labeled standard. Coelution so the C18:1 isomers is highlighted by the red circle.

what was shown previously is Chapter 3, with a few exceptions. The first is that the resolution obtained for the C18:1 isomers is completely lost with the addition of the labeled standard and the overloading causes relative standard deviations to be three times higher than those reported in Chapter 3. Additionally, if a split ratio or a dilution of the isotope labeled standard are used to try and resolve the co-elution issue, the less concentrated analytes of interest are lost. Thus, further optimization of this methodology is required. Pancreatic -cells In the work presented in Chapters 2 and 3, INS-1 cells, clonal -cells, were used for the analysis of -cells at varying concentrations of glucose. This is a common practice in metabolomics research due to the ease of obtaining significant quantities of INS-1 cells compared to isolating islets from animals. However, some questions have been raised in 123

the literature about the homogeneity of clonal -cells as well as the possibility of the metabolism being devoted primarily to cellular replication (a low frequency occurrence in postmitotic primary -cells).7 Thus, repetition of the experiments presented here with pancreatic islets isolated from mice would be beneficial in obtaining an improved knowledge of -cell metabolism. The increased detectability provided by GC GC will likely ease the number of islets required to obtain detectable amounts of metabolites. Therefore, with appropriate modification of the extraction procedures to account for isolated islets not being fixed to the cell plates, the metabolite profiling, metabolomic and fatty acid methyl ester analyses presented here could reasonably be repeated using isolated mice islets instead of clonal, INS-1, -cells. Sapropel Studies The pilot study of pyrolysis GC GC analysis of Mediterranean sapropels, presented in Chapter 4, further demonstrates the significant potential of this technique in the geochemical community. Optimization of the methods could be achieved by analyzing split ratio, column selection, and sample size. Additionally, implementation of an alternative modulation method may greatly benefit the analysis of higher molecular weight analytes. Through personal communication with the authors of a previous geochemical study8, we were made aware of limitations of the dual stage thermal modulator used in this work. For analytes in the C30 to C40 range, the liquid nitrogen cooled nitrogen cold jets are too cold and the compounds are not sufficiently released onto the second column. To perform modulation better suited for the higher boiling compounds present in sapropel samples, the solenoid valves that pulse the cold jets could be by-passed, and 124

instead a continuous cold jet flow could be run at a couple liters per minute (divided between the two jets). When the hot jets switch on, the cold flow would be blown away similar to other modulators used for GC GC. Alternatively, the experiment could be run in two parts. First, samples would be analyzed focusing on low-boiling components using an optimized version of the chromatography presented in Chapter 4 and the current modulator conditions. A second set of identical samples could then be run focusing on the high-boiling point compounds. In this instance, the liquid nitrogen would be omitted from the modulation process, and room temperature compressed nitrogen would be used for the cold jet. Environmental Implications The recent oil spill in the Gulf of Mexico, caused by the explosion of the Deepwater Horizon drilling rig on April 20, 2010, is the largest oil spill in US history, to date. It has been estimated that 4.4 million ( 20 %) barrels of oil, the equivalent of 7.0 105 m3, were released into the Gulf of Mexico.9 The environmental implications of such a spill are severe and will continue to be observed well into the future. Analysis of contaminated sediments (both terrestrial and sea) will certainly be of interest both environmentally and geochemically for years to come. Previous GC GC studies have evaluated the weathering of oil in previous spill locations and found the geochemical information obtained could be used to better understand biodegradation and phototransformation in the field and improve estimates of hydrocarbon concentrations in the water column.10 Thus, the methods described previously could lead to improved assessments of wildlife exposures to hydrocarbons after oil spills.10

125

The pyrolysis GC GC method presented in Chapter 4 could be adapted for the analysis of sediment from various underwater locations in the Gulf of Mexico. This method could further help improve the understanding of oil weathering after large scale spills. Additionally, because extraction is not necessary for pyrolysis analysis, the use of pyrolysis GC GC could significantly increase sample throughput. With higher sample throughput a greater number of geographical locations could be targeted and analyzed without additional time requirements, further increasing the information available about the weathering and environmental effects of the Gulf Coast oil spill. Overall Chromatographic Improvement The recent advancement in the Leco ChromaTOF software (v. 4.30) and the Pegasus modulator, which allows for what is known as variable modulation, may also prove to be beneficial in the analysis of complex samples such as those presented in Chapters 2 and 4. Variable modulation is demonstrated in Figure 5.2, a sample chromatogram obtained from Leco Separation Science. The modulation period in Figure 5.2 is increased from 3 to 4 to 5 seconds as the analysis is completed. This technique can be a significant time saver as the second dimension separation time can be tailored for portions of the chromatogram rather than just the component that elutes last from the second dimension. An increase of resolution in the first dimension is obtained by decreasing the modulation period in regions of the chromatogram that do not contain analytes which are highly retained in the second dimension.11 Additionally, variable modulation allows for the hot pulse time to be varied throughout the chromatographic analysis.

126

Figure 5.2 GC GC chromatogram using variable modulation. Notched appearance of the upper left portion of the chromatogram indicates where the modulation period was increased from 3 to 4 to 5 seconds. (Chromatogram obtained from Leco Separation Science and used with permission) 11

Prior to variable modulation, the hot pulse time was selected to provide the best overall second dimension peak shape across the chromatogram. The ability to decrease the hot pulse duration allows for increased trapping efficiency in the modulator in regions with more volatile analytes. Conversely, desorption efficiency of less volatile analytes is increased by utilizing hot pulses of increased duration.11 These new features will allow for improved chromatography in complex samples and thus have the potential to aid in the separation and detection of novel and or significant analytes that were previously identifiable even in GC GC.

127

References (1) (2) (3) Martins, E. F.; Miyasaka, C. K.; Newsholme, P.; Curi, R.; Carpinelli, A. R. Diabetes Metab. 2004, 30, 21-27. MacDonald, M. J.; Dobrzyn, A.; Ntambi, J.; Stoker, S. W. Arch. Biochem. Biophys. 2008, 470, 153-162. Persaud, S. J.; Muller, D.; Belin, V. D.; Kitsou-Mylona, I.; Asare-Anane, H.; Papadimitriou, A.; Burns, C. J.; Huang, G. C.; Amiel, S. A.; Jones, P. M. Diabetes 2007, 56, 197-203. Fischer, C. Biomed. Mass Spectrom. 1984, 11, 114-117. Gao, Z. G.; Yin, J.; Zhang, J.; Ward, R. E.; Martin, R. J.; Lefevre, M.; Cefalu, W. T.; Ye, J. P. Diabetes 2009, 58, 1509-1517. Li, J.; Yue, Y.; Hu, X.; Zhong, H. Anal. Chem. 2009, 81, 5080. Fernandez, C.; Fransson, U.; Hallgard, E.; Spegel, P.; Holm, C.; Krogh, M.; Warell, K.; James, P.; Mulder, H. J. Proteome Res. 2008, 7, 400-411. Ventura, G. T.; Kenig, F.; Reddy, C. M.; Frysinger, G. S.; Nelson, R. K.; Van Mooy, B.; Gaines, R. B. Org. Geochem. 2008, 39, 846-867. Crone, T. J.; Tolstoy, M. Science 2010, science.1195840. Arey, J. S.; Nelson, R. K.; Reddy, C. M. Environ. Sci. Technol. 2007, 41, 57385746. Leco In Leco Applications Notes; Leco Corporation: St. Joseph, MI, 2010.

(4) (5) (6) (7) (8) (9) (10) (11)

128

Appendix A List of Peaks Identified as changing by Fisher Ratio Analysis from 3 mM glucose to 7 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas

129

Peak Area Fisher Ratio 1422.7 15842 9400.7 5047.9 3852.4 3628.6 3462.3 3361.4 3356.5 3000.8 2868 2542 2244.2 2030.3 1998.1 1925.7 1862 1792.1 1779 1748.8 1674.2 1640.9 1528.2 1511.1 1482.4 Direction of Change KEGG ID C00134 C00392 C00122 C00064 C00149 C00246 C16639 C03826 C00009 C00408 C00124 C08322 C00049 C00880 C00489 C01685 C00037 C00180 C00249 C02237 C00163 C00041 C00116 C00095 C06424 7 mM - 3 mM -3.89E+05 1.60E+05 8.12E+05 1.40E+07 1.84E+06 -5.03E+06 9.07E+05 3.32E+04 -1.30E+07 2.97E+06 4.15E+05 -5.03E+04 -2.57E+06 -1.37E+05 4.32E+04 1.78E+03 2.06E+06 -7.09E+05 -8.80E+05 1.31E+05 1.53E+05 -2.43E+04 3.57E+04 -1.22E+05 -5.47E+04

Software Identification 1,4-Butanediamine, N,N,N',N'-tetrakis(trimethylsilyl)MANNITOL TMS 2-Butenedioic acid (E)-, bis(trimethylsilyl) ester Glutamine, tris(trimethylsilyl)Malic acid, tris(trimethylsilyl) ester Butanoic acid, 4-[bis(trimethylsilyl)amino]-, trimethylsilyl ester D-Ribofuranose, 1,2,3,5-tetrakis-O-(trimethylsilyl)Erythro-Pentonic acid, 2-deoxy-3,4,5-tris-O-(trimethylsilyl)-, trimethylsilyl ester PHOSPHORIC ACID,O,O,O-TMS 2-Piperidinecarboxylic acid, tert-butyldimethylsilyl ester, (DL)-

3 mM 2.72E+06 3.61E+04 2.43E+05 3.08E+06 5.26E+05 7.14E+06 1.54E+06 2.82E+04 1.61E+07 5.47E+05 4.45E+03 1.00E+05 9.05E+06 2.00E+05 1.61E+04 4.93E+03 3.60E+06 2.09E+06 4.21E+06 2.95E+05 1.35E+05 3.45E+04 3.15E+03 1.43E+05 1.85E+05

7 mM 2.33E+06 1.96E+05 1.05E+06 1.71E+07 2.37E+06 2.11E+06 2.45E+06 6.14E+04 3.09E+06 3.52E+06 4.20E+05 4.99E+04 6.48E+06 6.34E+04 5.93E+04 6.70E+03 5.66E+06 1.38E+06 3.33E+06 4.26E+05 2.88E+05 1.02E+04 3.89E+04 2.13E+04 1.31E+05

130

GALACTOSE MEOX1 TMS 9-Tetradecenoic acid, trimethylsilyl ester L-Aspartic acid, N-(trimethylsilyl)-, bis(trimethylsilyl) ester Galactonic acid, 2,3,4,5,6-pentakis-O-(trimethylsilyl)-, trimethylsilyl ester Pentanedioic acid, 2-(methoxyimino)-, bis(trimethylsilyl) ester Ribonic acid, 2,3,4,5-tetrakis-O-(trimethylsilyl)-, trimethylsilyl ester GLYCINE,N,N,O-TMS Benzoic acid trimethylsilyl ester Palmitic Acid Pyroglutamic acid Propanoic acid, 2-(methoxyimino)-, trimethylsilyl ester ALANINE,N,O-TMS Trimethylsilyl ether of glycerol FRUCTOSE MEOX1 5TMS Tetradecanoic acid, trimethylsilyl ester

RHAMNOSE MEOX2 4TMS XYLULOSE MEOX 4TMS Uridine, 2',3',5'-tris-O-(trimethylsilyl)4-AMINOBUTYRIC ACID 3TMS trans, trans-Farnesol, trimethylsilyl ether Myo-Inositol, 1,2,3,4,5,6-hexakis-O-(trimethylsilyl)Dodecanoic acid, ethyl ester L-Norvaline, N-(trimethylsilyl)-, trimethylsilyl ester Urea, N,N'-bis(trimethylsilyl)SUCROSE TMS Tyrosine, O-trimethylsilyl-, trimethylsilyl ester Acetic acid, bis[(trimethylsilyl)oxyl]-, trimethylsilyl ester Trimethylsiloxy(trimethylsilyl)proline Arabinofuranose, 1,2,3,5-tetrakis-O-(trimethylsilyl)Alanine, phenyl-, trimethylsilyl ester, dlPropanoic acid, 2-[(trimethylsilyl)oxy]-, trimethylsilyl ester Octanoic acid, trimethylsilyl ester Sebacic acid, bis(trimethylsilyl) ester L-Cysteine, N,S-bis(trimethylsilyl)-, trimethylsilyl ester ERYTHROSE MEOX1 3TMS Kaur-15-ene, (5,9,10)CITRIC ACID TMS Pentadecane HYDROXYBUTANOIC ACID,O,O-TMS Tris(trimethylsilyl)hydroxylamine D-Ribose, 2,3,4-tris-O-(trimethylsilyl)-, O-methyloxime, 5-[bis(trimethylsilyl) phosphate] Azulene, 1,2,3,4,5,6,7,8-octahydro-1,4-dimethyl-7-(1-methylethylidene)-, (1S-cis)GLUCOSE MEOX2 5TMS

1472.5 1422.9 1409.2 1398.3 1384.6 1371.2 1367.2 1345.4 1340.8 1336.5 1302.2 1281.7 1262.2 1261.2 1248.6 1207.2 1204.2 1203 1179.4 1177.7 1174.5 1164.7 1156.2 1155.8 1154.6 1123 1121.6 1118.9

C00507 C00312 C00299 C00334 C01126 C00137 C02679 C01826 C00086 C00089 C00082 C00033 C00148 C06115 C00079 C00022 C06423 C08277 C00097 C01796 C06090 C00158 C08388 C00989 C00192 C00117 C13392 C00031

1.45E+04 3.22E+03 -4.17E+04 1.26E+03 1.97E+04 -5.39E+05 -6.28E+03 -2.67E+05 -5.05E+05 -1.54E+04 1.47E+05 8.42E+04 2.50E+04 -1.79E+04 -3.08E+05 -6.56E+03 -1.09E+04 3.50E+03 -2.24E+04 -3.15E+04 -4.36E+04 3.98E+06 -2.50E+04 9.95E+03 1.21E+04 -2.65E+04 5.72E+03 2.13E+06

3.10E+04 3.38E+03 9.51E+04 3.60E+04 2.91E+04 3.21E+06 1.05E+04 3.21E+05 1.27E+06 2.88E+04 7.58E+05 4.78E+04 2.78E+04 1.77E+05 6.39E+05 1.88E+04 6.15E+04 2.64E+03 2.32E+05 7.38E+04 1.78E+05 0.00E+00 4.74E+04 2.64E+04 4.37E+04 1.07E+05 1.31E+04 1.68E+06

4.55E+04 6.60E+03 5.34E+04 3.73E+04 4.88E+04 2.67E+06 4.26E+03 5.40E+04 7.70E+05 1.34E+04 9.05E+05 1.32E+05 5.28E+04 1.59E+05 3.31E+05 1.22E+04 5.06E+04 6.14E+03 2.09E+05 4.24E+04 1.35E+05 3.98E+06 2.24E+04 3.64E+04 5.59E+04 8.04E+04 1.88E+04 3.81E+06

131

DIETHANOLAMINE,N,O,O-TMS ACONITIC ACID 3TMS ISOLEUCINE,N,O-TMS D-Glucuronic acid, 2,3,4,5-tetrakis-O-(trimethylsilyl)-, trimethylsilyl ester N,O,O-Tris(trimethylsilyl)-L-threonine Phenylethanolamine triTMS 9,12-Octadecadienoic acid (Z,Z)-, trimethylsilyl ester l-Valine, trimethylsilyl ester TYRAMINE,N,O-TMS

1116.5 1115.4 1115.1 1113.9 1108.4 1108 1089.6 1077 1071

C06772 C00417 C00407 C16245 C00188 C02735 C01595 C00183 C00483

-7.25E+03 5.69E+03 1.76E+05 3.57E+03 3.24E+03 -1.23E+04 1.07E+05 -4.48E+04 1.47E+04

2.99E+04 9.07E+03 7.65E+05 1.37E+04 1.71E+04 3.51E+04 1.02E+05 8.25E+04 5.34E+04

2.26E+04 1.48E+04 9.41E+05 1.73E+04 2.04E+04 2.28E+04 2.09E+05 3.77E+04 6.82E+04

132

Appendix B List of Peaks Identified as changing by Fisher Ratio Analysis from 3 mM glucose to 17 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas

133

Peak Area Fisher Ratio 9683 7943.6 5809.5 4148.8 4120.1 3996.4 3920.7 3807.2 3702.4 3648.4 3575.8 3464.8 3338.2 3308.7 3239.6 3198.3 3023.3 2855.1 2841.2 2807.1 2763.4 2611.7 2609.5 2579.4 Direction of Change

Software Identification 2-Butenedioic acid (E)-, bis(trimethylsilyl) ester XYLOSE MEOX2 4TMS GALACTOSE MEOX1 TMS 2-Propenoic acid, 2-[(trimethylsilyl)oxy]-, anhydride with bis(trimethylsilyl) hydrogen phosphate MANNITOL TMS D-Ribofuranose, 1,2,3,5-tetrakis-O-(trimethylsilyl)Glutamine, tris(trimethylsilyl)Malic acid, tris(trimethylsilyl) ester Tyrosine, O-trimethylsilyl-, trimethylsilyl ester Erythro-Pentonic acid, 2-deoxy-3,4,5-tris-O-(trimethylsilyl)-, trimethylsilyl ester trans-9-Octadecenoic acid, trimethylsilyl ester ISOLEUCINE,N,O-TMS L-Aspartic acid, N-(trimethylsilyl)-, bis(trimethylsilyl) ester SERINE,N,O,O-TMS XYLITOL 5TMS trans, trans-Farnesol, trimethylsilyl ether N,O,O-Tris(trimethylsilyl)-L-threonine LEUCINE, N,O-TMS SUCROSE TMS Propanoic acid, 2-oxo-3-(trimethylsilyl)-, trimethylsilyl ester 2-Piperidinecarboxylic acid, tert-butyldimethylsilyl ester, (DL)3-OXOGLUTARIC ACID MEOX1 TMS Pyroglutamic acid XYLULOSE MEOX 4TMS

KEGG ID C00122 C01394 C00124 C00511 C00392 C16639 C00064 C00149 C00082 C03826 C01712 C16434 C00049 C00065 C00379 C01126 C00188 C00123 C00089 C00163 C00408 C00026 C02237 C00312

17 mM - 3 mM 1.43E+06 3.70E+04 -1.48E+06 1.11E+04 2.92E+05 1.09E+06 5.41E+06 1.72E+05 1.12E+06 5.24E+04 -2.01E+05 9.94E+05 -3.71E+06 1.09E+06 1.48E+04 2.84E+04 2.70E+05 2.75E+06 -1.66E+04 -3.89E+04 2.76E+05 4.02E+05 9.19E+04 2.23E+06

3 mM 2.43E+05 2.07E+04 1.52E+06 3.45E+02 7.75E+03 1.54E+06 3.63E+05 6.31E+04 7.58E+05 2.65E+04 6.83E+05 7.65E+05 8.13E+06 1.23E+06 1.24E+03 2.91E+04 4.97E+05 2.95E+06 2.62E+04 1.07E+05 1.43E+05 1.05E+06 1.41E+04 3.12E+06

17 mM 1.67E+06 5.78E+04 4.57E+04 1.15E+04 3.00E+05 2.63E+06 5.77E+06 2.35E+05 1.88E+06 7.89E+04 4.82E+05 1.76E+06 4.42E+06 2.32E+06 1.60E+04 5.75E+04 7.67E+05 5.69E+06 9.63E+03 6.79E+04 4.19E+05 1.45E+06 1.06E+05 5.35E+06

134

Palmitic Acid Tetradecanoic acid, trimethylsilyl ester ALANINE,N,O-TMS HYDROXYBUTANOIC ACID,O,O-TMS Uridine, 2',3',5'-tris-O-(trimethylsilyl)RHAMNOSE MEOX2 4TMS Butanal, 2,3,4-tris[(trimethylsilyl)oxy]-, (R*,R*)L-Valine, N-(trimethylsilyl)-, trimethylsilyl ester Ribonic acid, 2,3,4,5-tetrakis-O-(trimethylsilyl)-, trimethylsilyl ester ERYTHROSE MEOX2 TMS -DL-Arabinopyranose, 1,2,3,4-tetrakis-O-(trimethylsilyl)D-Xylose, tetrakis(trimethylsilyl)Benzoic acid trimethylsilyl ester Pentanedioic acid, 2-(methoxyimino)-, bis(trimethylsilyl) ester OXALIC ACID TMS Acetic acid, bis[(trimethylsilyl)oxyl]-, trimethylsilyl ester LAURIC ACID TMS Galactonic acid, 2,3,4,5,6-pentakis-O-(trimethylsilyl)-, trimethylsilyl ester 4-AMINOBUTYRIC ACID 3TMS Lyxose, tetra-(trimethylsilyl)-ether Myo-Inositol, 1,2,3,4,5,6-hexakis-O-(trimethylsilyl)D-Fructose, 1,3,4,5,6-pentakis-O-(trimethylsilyl)Alanine, phenyl-, trimethylsilyl ester, dlArachidonic acid, trimethylsilyl ester Azelaic acid, bis(trimethylsilyl) ester O,O,O'-Tris-trimethylsilylmalonate Glyoxylic acid, di-TMS Azulene, 1,2,3,4,5,6,7,8-octahydro-1,4-dimethyl-7-(1-methylethylidene)-, (1Scis)-

2464.2 2298.7 2174.8 2156.1 2155.5 2129.8 2117.7 2041.7 1975 1950.9 1911.8 1897.4 1886.9 1814.6 1739.9 1722.5 1694.1 1653.1 1638.9 1618 1610.7 1570.7 1536.2 1475.4 1447.2 1427.3 1424.2 1408.3

C00249 C06424 C00041 C01089 C00299 C00507 C01412 C00183 C01685 C01796 C00259 C00181 C00180 C00489 C00209 C00033 C02679 C00880 C00334 C01508 C00137 C00095 C00079 C00219 C08261 C00383 C00048 C13392

9.32E+03 -7.76E+05 -6.34E+04 2.29E+05 1.09E+05 -5.35E+04 2.93E+04 2.68E+04 1.62E+06 5.22E+03 3.24E+06 1.77E+05 9.18E+04 -8.12E+05 -1.29E+03 3.08E+06 -2.09E+04 -2.84E+04 7.21E+03 -2.24E+04 8.70E+04 -2.07E+05 6.64E+03 9.46E+03 9.46E+03 1.93E+04 5.28E+03 5.81E+04

3.47E+03 4.21E+06 1.85E+05 6.29E+04 1.44E+04 9.51E+04 3.10E+04 2.21E+04 2.60E+06 4.93E+03 3.55E+03 2.63E+05 7.98E+04 2.09E+06 5.88E+03 3.60E+05 4.78E+04 9.08E+04 3.71E+03 4.06E+04 7.95E+04 3.21E+06 1.12E+04 3.54E+04 3.54E+04 5.44E+04 5.16E+03 8.55E+04

1.28E+04 3.43E+06 1.22E+05 2.92E+05 1.24E+05 4.16E+04 6.03E+04 4.89E+04 4.22E+06 1.01E+04 3.25E+06 4.39E+05 1.72E+05 1.27E+06 4.59E+03 3.44E+06 2.69E+04 6.24E+04 1.09E+04 1.83E+04 1.67E+05 3.00E+06 1.78E+04 4.49E+04 4.49E+04 7.37E+04 1.04E+04 1.44E+05

135

1-Octanol, 2,2-dimethylTrimethylsiloxy(trimethylsilyl)proline Octadecanoic acid, trimethylsilyl ester Arabinofuranose, 1,2,3,5-tetrakis-O-(trimethylsilyl)ORNITHINE,N,N,N',O-TMS Gluconic acid, 2-methoxime, tetra(trimethylsilyl)-, trimethylsilyl ester UREA 2TMS -Elemene ASPARAGINE,N,N,O-TMS Octanedioic acid, bis(trimethylsilyl) ester MALTOSE MEOX2 TMS Kaur-15-ene, (5,9,10)Hexanoic acid, trimethylsilyl ester Tridecane Glycine, N-formyl-, trimethylsilyl ester Galacturonic acid, pentakis(trimethylsilyl)Decanal, O-methyloxime GLYCEROL 3TMS ERYTHROSE MEOX1 3TMS 3-(Trimethylsiloxy)cholest-5-ene -D-Fructofuranose, 2,3,4,6-tetrakis-O-(trimethylsilyl)-, bis(trimethylsilyl) phosphate Pentadecane NORLEUCINE,O-TMS TYRAMINE,N,O-TMS Acetoacetic acid, bis(trimethylsilyl)- deriv. Heptanedioic acid, bis(trimethylsilyl) ester D-Glucuronic acid, 2,3,4,5-tetrakis-O-(trimethylsilyl)-, trimethylsilyl ester

1407.4 1400.5 1400.1 1395.8 1363.4 1301.7 1286.6 1273.3 1265.3 1265 1247.3 1241 1212.1 1175.7 1165 1160 1158.6 1150.5 1148 1139 1138.8 1138 1137.7 1133.7 1095.6 1090.5 1089.2

C00756 C00148 C01530 C06115 C00077 C00257 C00086 C17094 C00152 C08278 C00208 C06090 C01585 C13834 C00037 C00333 C12307 C00116 C01796 C05416 C03267 C08388 C01933 C00483 C00164 C02656 C16245

1.84E+04 -2.09E+05 5.06E+04 2.88E+03 -2.19E+04 -1.14E+05 7.46E+05 -2.67E+04 4.03E+03 9.54E+03 2.29E+03 6.31E+03 -7.32E+04 -8.26E+03 4.49E+06 1.43E+03 6.35E+03 1.05E+04 -5.47E+04 3.81E+04 2.02E+04 -8.36E+04 4.85E+04 -1.17E+04 3.40E+03 3.64E+03 8.99E+03

1.51E+03 9.35E+05 2.60E+04 2.70E+03 1.77E+05 3.02E+05 75710.785 31646.636 8265.44 3.59E+04 8.24E+03 1.77E+04 2.40E+05 2.78E+04 7.31E+05 1.35E+04 1.64E+04 3.38E+04 3.30E+05 5.17E+04 4.73E+04 1.38E+05 1.66E+05 1.63E+05 6.73E+03 3.65E+04 9.59E+03

1.99E+04 7.27E+05 7.66E+04 5.58E+03 1.55E+05 1.89E+05 821237.285 4927.352 12295.035 4.55E+04 1.05E+04 2.40E+04 1.66E+05 1.95E+04 5.22E+06 1.50E+04 2.27E+04 4.42E+04 2.75E+05 8.99E+04 6.75E+04 5.43E+04 2.15E+05 1.52E+05 1.01E+04 4.02E+04 1.86E+04

136

Appendix C List of Peaks Identified as changing by Fisher Ratio Analysis from 7 mM glucose to 17 mM glucose with KEGG identifications, direction of change, Fisher Ratios and Peak Areas

137

Peak Area Fisher Ratio 1339 6406.5 4722.9 4232.4 4006.3 3664.2 3521.4 3515.9 3069.2 2992.7 2961 2806.9 2508.1 2337.8 2308.2 2217.8 2137.4 2064 2036.4 1981 1952.6 1928 1853.5 1837.5 1771.6 Direction of Change KEGG ID C00299 C01394 C00246 C01796 C00064 C00049 C00124 C00198 C00163 C00123 C00065 C00188 C00041 C00122 C00082 C16434 C00159 C00009 C00880 C00392 C00037 C00219 C00089 C00048 C01089 7 mM - 3 mM -1.70E+04 3.43E+04 -7.91E+05 3.38E+06 -1.15E+07 -3.14E+06 -4.03E+06 2.43E+03 1.86E+06 2.36E+06 9.65E+05 2.32E+05 -9.79E+04 6.19E+05 9.72E+05 8.18E+05 -1.96E+04 4.18E+06 -1.86E+05 1.71E+05 -7.43E+04 1.34E+04 -3.74E+03 7.38E+04 9.12E+04

Software Identification Uridine, 2',3',5'-tris-O-(trimethylsilyl)XYLOSE MEOX2 4TMS Butanoic acid, 4-[bis(trimethylsilyl)amino]-, trimethylsilyl ester ERYTHROSE MEOX2 TMS Glutamine, tris(trimethylsilyl)L-Aspartic acid, N-(trimethylsilyl)-, bis(trimethylsilyl) ester GALACTOSE MEOX1 TMS Gluconic acid, -lactone, 5-methoximine, tri(trimethylsilyl)Propanoic acid, 2-[(trimethylsilyl)oxy]-, trimethylsilyl ester LEUCINE, N,O-TMS

3 mM 5.54E+04 2.35E+04 1.62E+06 1.65E+03 1.61E+07 5.78E+06 8.11E+06 6.22E+03 3.35E+06 3.33E+06 1.23E+06 5.35E+05 1.59E+05 1.05E+06 9.05E+05 9.41E+05 2.84E+04 1.16E+06 2.08E+05 1.96E+05 1.70E+05 3.15E+04 1.34E+04 6.98E+04 3.79E+04

7 mM 3.84E+04 5.78E+04 8.29E+05 3.38E+06 4.59E+06 2.64E+06 4.08E+06 8.65E+03 5.21E+06 5.69E+06 2.20E+06 7.67E+05 6.07E+04 1.67E+06 1.88E+06 1.76E+06 8.84E+03 5.34E+06 2.15E+04 3.66E+05 9.61E+04 4.49E+04 9.63E+03 1.44E+05 1.29E+05

SERINE,N,O,O-TMS N,O,O-Tris(trimethylsilyl)-L-threonine ALANINE,N,O-TMS 2-Butenedioic acid (E)-, bis(trimethylsilyl) ester Tyrosine, O-trimethylsilyl-, trimethylsilyl ester ISOLEUCINE,N,O-TMS MANNOSE MEOX TMS PHOSPHORIC ACID,O,O,O-TMS Galactonic acid, 2,3,4,5,6-pentakis-O-(trimethylsilyl)-, trimethylsilyl ester MANNITOL TMS GLYCINE,N,N,O-TMS Arachidonic acid, trimethylsilyl ester SUCROSE TMS Glyoxylic acid, di-TMS HYDROXYBUTANOIC ACID,O,O-TMS

138

Azelaic acid, bis(trimethylsilyl) ester Azulene, 1,2,3,4,5,6,7,8-octahydro-1,4-dimethyl-7-(1-methylethylidene)-, (1S-cis)Lyxose, tetra-(trimethylsilyl)-ether 9,12-Octadecadienoic acid (Z,Z)-, trimethylsilyl ester Galacturonic acid, pentakis(trimethylsilyl)Dodecanoic acid, ethyl ester Pentanedioic acid, 2-(methoxyimino)-, bis(trimethylsilyl) ester Pyroglutamic acid XYLULOSE MEOX 4TMS L-Valine, N-(trimethylsilyl)-, trimethylsilyl ester PYROGLUTAMIC ACID 2TMS 3-OXOGLUTARIC ACID MEOX1 TMS trans, trans-Farnesol, trimethylsilyl ether D-Ribose, 2,3,4-tris-O-(trimethylsilyl)-, O-methyloxime, 5-[bis(trimethylsilyl) phosphate] -DL-Arabinopyranose, 1,2,3,4-tetrakis-O-(trimethylsilyl)Acetic acid, bis[(trimethylsilyl)oxyl]-, trimethylsilyl ester Alanine, phenyl-, trimethylsilyl ester, dlTrimethylsilyl acetylsalicylate GLYCEROL 3TMS OXALIC ACID TMS Propanedioic acid, bis(trimethylsilyl) ester Heptanedioic acid, bis(trimethylsilyl) ester METHIONINE,N,O-TMS MALTOSE MEOX2 TMS PIPECOLIC ACID,O-TMS 2-Propenoic acid, 3-(4-methoxyphenyl)-, 2-ethylhexyl ester Octanoic acid, trimethylsilyl ester ETHYLENEGLYCOL,O,O-TMS

1743 1733.8 1677.6 1636.7 1612 1605.3 1604.7 1525.1 1508.7 1506.6 1492.6 1485.4 1478.1 1442.4 1441.3 1431 1411 1313.7 1274.2 1259.3 1256.5 1245 1241.5 1221.4 1218.9 1212.7 1212.4 1207.5

C08261 C13392 C01508 C01595 C08348 C02679 C00489 C02237 C00312 C00183 C01879 C00026 C01126 C00117 C00259 C00033 C00079 C01405 C00116 C00209 C00383 C02656 C00073 C00208 C00408 C00511 C06423 C05576

1.89E+04 -7.88E+03 1.21E+05 -5.73E+04 1.95E+04 7.52E+03 -2.88E+02 4.04E+05 6.19E+03 1.77E+04 -9.84E+05 5.10E+04 8.11E+03 -1.25E+04 1.12E+05 -5.98E+03 2.78E+05 -4.16E+05 1.41E+04 -5.30E+03 4.74E+04 9.35E+03 1.51E+05 2.94E+03 -2.27E+05 -7.09E+05 8.38E+03 1.55E+04

5.48E+04 1.39E+04 8.13E+04 2.39E+05 9.43E+03 4.26E+03 4.63E+03 1.25E+05 6.60E+03 2.50E+04 1.23E+06 5.93E+04 4.94E+04 8.04E+04 3.27E+05 4.09E+04 3.31E+05 4.32E+05 3.02E+04 8.39E+03 3.88E+04 3.08E+04 1.18E+05 7.59E+03 5.74E+05 8.77E+05 5.06E+04 2.03E+04

7.37E+04 6.00E+03 2.02E+05 1.81E+05 2.89E+04 1.18E+04 4.34E+03 5.29E+05 1.28E+04 4.27E+04 2.46E+05 1.10E+05 5.75E+04 6.79E+04 4.39E+05 3.49E+04 6.09E+05 1.64E+04 4.42E+04 3.09E+03 8.62E+04 4.02E+04 2.69E+05 1.05E+04 3.47E+05 1.68E+05 5.90E+04 3.58E+04

139

Kaur-15-ene, (5,9,10)ASPARAGINE,N,N,O-TMS Butanal, 2,3,4-tris[(trimethylsilyl)oxy]-, (R*,R*)RHAMNOSE MEOX2 4TMS Propanoic acid, 3-[(trimethylsilyl)oxy]-, trimethylsilyl ester D-Fructose, 1,3,4,5,6-pentakis-O-(trimethylsilyl)Aminomalonic acid, tris(trimethylsilyl)Acetoacetic acid, bis(trimethylsilyl)- deriv. Gluconic acid, 2-methoxime, tetra(trimethylsilyl)-, trimethylsilyl ester Decanal, O-methyloxime Malic acid, tris(trimethylsilyl) ester NORLEUCINE,O-TMS Octadecanoic acid, methyl ester 11-Eicosenoic acid, trimethylsilyl ester 3-(Trimethylsiloxy)cholest-5-ene Nonanoic acid, trimethylsilyl ester L-Cysteine, N,S-bis(trimethylsilyl)-, trimethylsilyl ester l-Valine, trimethylsilyl ester Phenylethanolamine triTMS Octanedioic acid, bis(trimethylsilyl) ester

1202.5 1198.9 1171.1 1160.2 1158.7 1154.5 1151 1144.2 1138.4 1132.3 1131.5 1111.1 1106.6 1103.2 1098.8 1086.2 1083.2 1074.6 1068.2 1067.4

C06090 C00152 C01412 C00507 C00022 C00095 C00872 C00164 C00257 C12307 C00497 C01933 C01530 C16526 C05416 C01601 C00097 C00183 C02735 C08278

7.04E+02 -6.54E+03 9.69E+03 1.48E+04 6.52E+04 9.35E+03 2.29E+05 3.47E+03 3.60E+03 8.20E+03 2.45E+05 -2.70E+03 1.02E+03 3.18E+02 1.13E+03 1.99E+04 1.27E+05 -4.34E+04 8.81E+03 9.59E+03

4.27E+03 2.18E+04 3.92E+04 4.55E+04 1.97E+05 1.07E+04 4.59E+05 6.66E+03 4.55E+03 1.15E+04 1.21E+06 2.12E+05 1.52E+04 1.45E+04 9.94E+04 6.21E+04 2.09E+05 4.43E+05 2.28E+04 3.59E+04

4.98E+03 1.53E+04 4.89E+04 6.03E+04 2.62E+05 2.00E+04 6.88E+05 1.01E+04 8.14E+03 1.97E+04 1.45E+06 2.09E+05 1.62E+04 1.48E+04 1.01E+05 8.20E+04 3.36E+05 4.00E+05 3.16E+04 4.55E+04

140

Appendix D List of Pathways Identified by Metscape Analysis from 3 mM Glucose to 7 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas

141

Average Peak Area Direction of Change

Reaction R00768 RE2754 RE2563 RE2649 RE2638 R03718 RE2650 R01176 RE1514 RE1576

Pathway Aminosugars metabolism Arachidonic acid metabolism Arachidonic acid metabolism Bile acid biosynthesis Bile acid biosynthesis Bile acid biosynthesis Biopterin metabolism Butanoate metabolism Di-unsaturated fatty acid beta-oxidation De novo fatty acid biosynthesis De novo fatty acid biosynthesis De novo fatty acid biosynthesis De novo fatty acid biosynthesis Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism

Seed glutamine glycerol glycerol propanoate glycine glycine phenylalanine butanoate linoleate myristic acid palmitate palmitate lauric acid d-fructose d-fructose d-fructose d-fructose d-glucose d-galactose d-galactose d-galactose d-galactose d-galactose d-galactose d-galactose

7 mM - 3 mM 1.40E+07 3.57E+04 1.53E+05 2.06E+06 -3.08E+05 -5.03E+06 1.07E+05 -5.47E+04 -8.80E+05 -6.28E+03 -1.22E+05

3 mM 3.08E+06 3.15E+03 1.35E+05 3.60E+06 639263.172 7.14E+06 102259.013 1.85E+05 4.21E+06 1.05E+04 1.43E+05

7 mM 1.71E+07 3.89E+04 2.88E+05 5.66E+06 330825.176 2.11E+06 208765.898 1.31E+05 3.33E+06 4.26E+03 2.13E+04

142

RN0034 R01706 RE1575 R00867 R00875 R00866 RE2783 RE1342 R03634 R01101 R05549 R01104 R01105 R01100 R01329

2.13E+06 4.15E+05

1684859.34 4.45E+03

3814592.53 4.20E+05

R01093 R01095 R02926 R01092 R01103 R01678 R01036 R03616 R00847 RE3507 R01352 R01041 R01350 R01351 R04452 R03038 R00369 RE1473 RE2642 R00945 R00371 R00366 R06171 R01221 RE2429 R00367 R00610

Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism

d-galactose d-galactose d-galactose d-galactose sucrose & d-galactose d-galactose glycerol glycerol glycerol glycerol glycerol glycerol glycerol glycerol acetate Alanine Alanine Alanine Alanine Glycine Glycine Glycine Glycine Glycine Glycine Glycine Glycine 2.06E+06 3.60E+06 5.66E+06 2.65E+04 -2.43E+04 14385.659 3.45E+04 40885.513 1.02E+04 3.57E+04 3.15E+03 3.89E+04 -1.54E+04 2.88E+04 1.34E+04

143

RE2427 R00751 R00565 R03425 RE2031 RE2428 R03654 R00611 RE2117 RE2111 R00220 R00704 R03663 R00710 R00227 R00235 R00317 R00316 R00216 R00214 R00200 R00344 R00006 R00703 R01699 R00210 R00196

Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis

Glycine Glycine Glycine Glycine Glycine Glycine Glycine Glycine Glycine Glycine pyruvate pyruvate threonine acetate acetate acetate acetate acetate malate & pyruvate malate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate 9.73E+04 143141.61 240399.287 1.84E+06 5.26E+05 2.37E+06 3.24E+03 2.65E+04 17137.327 14385.659 20378.485 40885.513 9.73E+04 143141.61 240399.287

144

R00711 R00209 R00014 RE2717 RE3632 RE1821 RE0596 R07056 R07057 R07064 RE3409 R07055 R03626 R07063 R03284 R02204 R02203 R02201 R02205 R00652 RE2811 R00896 R00782 R00891 RE2223 R01001

Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycosphingolipid metabolism Histidine metabolism Leukotriene metabolism Leukotriene metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism

pyruvate & acetate pyruvate pyruvate d-galactose Glycine Glycine Glycine linoleate linoleate linoleate linoleate linoleate linoleate linoleate Glycine Pipecolate Pipecolate Pipecolate Pipecolate Glycine cysteine cysteine cysteine & pyruvate cysteine cysteine cysteine 9.73E+04 143141.61 240399.287 2.06E+06 -2.24E+04 3.60E+06 231698.474 5.66E+06 209254.529 2.06E+06 2.13E+05 3.60E+06 3.03E+05 5.66E+06 5.16E+05 1.07E+05 102259.013 208765.898 4.15E+05 2.06E+06 2.06E+06 4.45E+03 3.60E+06 3.60E+06 4.20E+05 5.66E+06 5.66E+06

145

R00892 R03650 R00893 R03105 R01049 R01057 R01641 R01054 R06590 R01056 R01051 RE0124 R03331 R01186 R01187 RE3273 R01185 R01194 R01184 R00831 R00830 R01354 RE2067 RE2068 R02422 R01083 R01135 R01072

Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Porphyrin metabolism Porphyrin metabolism Propanoate metabolism Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Purine metabolism Purine metabolism Purine metabolism Purine metabolism

cysteine cysteine cysteine pyruvate ribose-5-phosphate ribose-5-phosphate ribose-5-phosphate ribose-5-phosphate ribose-5-phosphate ribose-5-phosphate ribose-5-phosphate d-gluconic acid glycerol myo-inositol myo-inositol myo-inositol myo-inositol myo-inositol myo-inositol Glycine Glycine propanoate glycerol glycerol urea fumarate aspartate glutamine -5.05E+05 8.12E+05 -2.57E+06 1.40E+07 1.27E+06 2.43E+05 9.05E+06 3.08E+06 7.70E+05 1.05E+06 6.48E+06 1.71E+07 1.53E+05 3.57E+04 1.35E+05 3.15E+03 2.88E+05 3.89E+04 2.06E+06 3.60E+06 5.66E+06 3.57E+04 -5.39E+05 3.15E+03 3.21E+06 3.89E+04 2.67E+06 -2.65E+04 106873.904 80412.779

146

R01231 R01397 R00575 R00963 R00967 R01876 R02327 R00968 R02097 R00970 R01549 R01880 R02332 R01878 R00964 R01274 R00408 R00342 R02164 R00412 R01082 R00362 R01325 R00351 R01324 R00352

Purine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Saturated fatty acids beta-oxidation TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle

glutamine aspartate glutamine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine palmitate fumarate malate fumarate fumarate malate & fumarate citrate citrate &cis-aconitate citrate citrate citrate 5.69E+03 9069.32 14762.068 3.98E+06 0 3975268.33 -8.80E+05 8.12E+05 1.84E+06 4.21E+06 2.43E+05 5.26E+05 3.33E+06 1.05E+06 2.37E+06 -2.57E+06 1.40E+07 -4.17E+04 9.05E+06 3.08E+06 9.51E+04 6.48E+06 1.71E+07 5.34E+04

147

R01900 RE1915 R01364 R01795 R00694 RE1465 R00689 R00698 R00692 R03660 R00699 RE1919 R02383 R02382 RE2124 RE2781 R00729 RE0026 R02078 R01815 RE0941 R02918 R00736 R03539

TCA cycle Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism

cis-aconitate cysteine fumarate phenylalanine & tyrosine phenylalanine phenylalanine & tyrosine phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine pyruvate tyramine tyramine tyramine tyramine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine & tyramine tyrosine 1.47E+05 758488.51 905112.62 9.73E+04 1.47E+04 143141.61 53443.233 240399.287 68159.058 -3.08E+05 639263.172 330825.176 -2.24E+04 8.12E+05 231698.474 2.43E+05 209254.529 1.05E+06

148

RE3062 R00734 R00731 R01920 RE1537 R01151 R01154 R00670 R01157 R00258

Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine

tyrosine tyrosine tyrosine putrescine putrescine putrescine putrescine putrescine putrescine alanine urea fumarate aspartate aspartate aspartate aspartate aspartate aspartate & glutamine -2.57E+06 1.40E+07 9.05E+06 3.08E+06 6.48E+06 1.71E+07 -2.43E+04 -5.05E+05 8.12E+05 3.45E+04 1.27E+06 2.43E+05 1.02E+04 7.70E+05 1.05E+06 -3.89E+05 2.72E+06 2.33E+06

149

R00551 R01086 R03647 R00357 R00488 R00355 R01954 R00578

R00489

aspartate

R03421 R00526 R05577 R02549 R01987 R00261 R01648 R01986 R01989

R00256 R03652 R00253 R00497 R04951 R00135 R01248 R01251 R01252 R03661

Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate,

aspartate aspartate aspartate 4-aminobutanoate 4-aminobutanoate 4-aminobutanoate 4-aminobutanoate 4-aminobutanoate 4-aminobutanoate glutamine glutamine glutamine glycine glycine proline proline proline proline proline 2.50E+04 27803.7 52777.33 2.06E+06 3.60E+06 5.66E+06 1.26E+03 3.60E+04 3.73E+04

150

aspartate and asparagine R01644 R00894 R02743 R00899 R00257 R02197 R03656 R02199 R02198 R03665 R01214 R04230 R04233 Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Vitamin B3 (nicotinate and nicotinamide) metabolism Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Vitamin B5 - CoA biosynthesis from pantothenate Vitamin B5 - CoA biosynthesis from pantothenate 4-hydroxybutanoate cysteine cysteine cysteine glutamine isoleucine isoleucine isoleucine isoleucine valine valine cysteine cysteine -2.24E+04 231698.474 209254.529 -4.48E+04 82483.602 37692.497 1.40E+07 1.76E+05 3.08E+06 765275.077 1.71E+07 941063.876 9.95E+03 -2.24E+04 26409.364 231698.474 36359.44 209254.529

151

Appendix E List of Pathways Identified by Metscape Analysis from 3 mM glucose to 17 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas

152

Average Peak Area Direction of Change

Reaction R00768 RE3647 R07052 RE3650 R07046 RE3286 R07050 R07041 R01317 RE3649

Pathway Aminosugars metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism

Seed glutamine arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate glycerol

17 mM - 3 mM 5.41E+06 9.46E+03

3 mM 3.63E+05 3.54E+04

17 mM 5.77E+06 4.49E+04

153

R01596 RE3651 RE3648 RE3653 RE3288 RE3287 RE3289 RE3646 R07054 R01593 R07051 RE3042 R07048 RE3654 RE2563

1.05E+04

33784.016

44244.974

RE2754 RE2649 RE2638 R03718 RE2649 RE2650 R01357 R00410

Arachidonic acid metabolism Bile acid biosynthesis Bile acid biosynthesis Bile acid biosynthesis Bile acid biosynthesis Biopterin metabolism Butanoate metabolism Butanoate metabolism

glycerol glycine glycine glycine propanoate tyrosine acetoacetate acetoacetate 3hydroxybutanoi c acid lauric acid stearate myristic acid palmitate palmitate mannose d-fructose d-fructose d-fructose d-fructose d-galactose d-galactose d-galactose d-galactose & sucrose -1.66E+04 2.62E+04 9.63E+03 d-galactose d-galactose d-galactose -1.48E+06 1.52E+06 4.57E+04 6.64E+03 1.12E+04 1.78E+04 1.42E+05 1.12E+06 3.40E+03 1.21E+05 7.58E+05 6727.541 2.62E+05 1.88E+06 10130.481 4.49E+06 730562.804 5222233.54

R01361 RE1575 RE0344 RE1576 RN0034 R01706 R01326 R00875 R00866 R00867 RE2783 R03634 R01105 R01100 R01103 R01678 R01101 R05549

Butanoate metabolism De novo fatty acid biosynthesis De novo fatty acid biosynthesis De novo fatty acid biosynthesis De novo fatty acid biosynthesis De novo fatty acid biosynthesis Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism

1.09E+05 -2.84E+04 2.88E+03 -6.34E+04 -7.76E+05

1.44E+04 9.08E+04 2.70E+03 1.85E+05 4.21E+06

1.24E+05 6.24E+04 5.58E+03 1.22E+05 3.43E+06

154

R01104 R01329 R01093 R01095 R02926 R01092 R04452 RE3299 RE3298 RE3301 R00847 R01036 R01041 R01350 R01351 R01352 R03616 RE3507 R00945 R00366 R00751 R00582 R00220 R00585 R03662

Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism

d-galactose d-galactose d-galactose d-galactose d-galactose d-galactose acetate serine serine serine glycerol glycerol glycerol glycerol glycerol glycerol glycerol glycerol glycine & serine glycine & glyoxylate glycine & threonine serine serine serine serine 1.09E+06 1.23E+06 2.32E+06 1.05E+04 33784.016 44244.974 -2.09E+04 1.09E+06 4.78E+04 1.23E+06 2.69E+04 2.32E+06

155

R00717 R00475 R00465 R03038 RE1473 RE2642 R00369 RE2031 R03663 R00367 R00371 R00565 R00610 R00611 R01221 R03425 R03654 R06171 RE2111 RE2117 RE2427 RE2428 RE2429 R00711 R00316 R00227 R00710 R00235

Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis

glyoxylate glyoxylate glyoxylate alanine alanine alanine alanine alanine threonine glycine glycine glycine glycine glycine glycine glycine glycine glycine glycine glycine glycine glycine glycine acetate acetate acetate acetate acetate

5.81E+04

8.55E+04

1.44E+05

2.29E+05

6.29E+04

2.92E+05

2.70E+05 4.49E+06

4.97E+05 730562.804

7.67E+05 5222233.54

156

-2.09E+04

4.78E+04

2.69E+04

R00317 R00214 R00216 RE2717 R01281 RE3632 RE0596 RE1821 R01595 RE3417 RE3420 R03284 R02204 R02203 R02201 R02205 R00652 R01289 R00891 R00590 R01290 RE0324 RE0124 R01187 R01186 RE3273 R01185 R01184

Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycosphingolipid metabolism Glycosphingolipid metabolism Histidine metabolism Leukotriene metabolism Leukotriene metabolism Leukotriene metabolism Linoleate metabolism Linoleate metabolism Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Omega-6 fatty acid metabolism Pentose phosphate pathway Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism

acetate malate malate d-galactose serine glycine glycine glycine arachidonate azelaic acid azelaic acid glycine pipecolate pipecolate pipecolate pipecolate glycine serine serine serine serine arachidonate d-gluconic acid m yo-inositol m yo-inositol m yo-inositol m yo-inositol m yo-inositol 9.46E+03 4.65E+03 -2.07E+05 3.54E+04 3496.041 3.21E+06 4.49E+04 8144.155 3.00E+06 1.09E+06 1.23E+06 2.32E+06 4.49E+06 4.02E+05 730562.804 1.05E+06 5222233.54 1.45E+06 9.46E+03 1.93E+04 3.54E+04 5.44E+04 4.49E+04 7.37E+04 -1.48E+06 1.09E+06 4.49E+06 4.49E+06 1.52E+06 1.23E+06 730562.804 730562.804 4.57E+04 2.32E+06 5222233.54 5222233.54 1.72E+05 6.31E+04 2.35E+05

157

R01194 R03331 R01354 RE1978 RE3455 R01590 RE3452 RE3458 RE3449 RE2068 RE2067 R01072 R01231 R01135 R01083 R02422 R01876 R02327 R02097 R00970 R01549 R01880 R02332 R00964 R00963 R00967 R00968 R01878

Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Propanoate metabolism Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Purine metabolism Purine metabolism Purine metabolism Purine metabolism Purine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism

m yo-inositol glycerol propanoate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate glycerol glycerol glutamine glutamine asparate fumarate urea uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine -3.71E+06 1.43E+06 7.46E+05 -5.35E+04 8.13E+06 2.43E+05 75710.785 9.51E+04 4.42E+06 1.67E+06 821237.285 4.16E+04 5.41E+06 3.63E+05 5.77E+06 1.05E+04 33784.016 44244.974 1.05E+04 1.42E+05 9.46E+03 33784.016 1.21E+05 3.54E+04 44244.974 2.62E+05 4.49E+04

158

R00575 R01397 R01274 R00342 R00362 R00412 R01082 R00408 R02164 R00268 RE0361 R00621 R00709 R01795 RE3062 RE0026 R00734 R02078 RE1465 R00736 RE0941 R01815 R00729 R02918 R03539 R00731 R01364 R00692

Pyrimidine metabolism Pyrimidine metabolism Saturated fatty acids beta-oxidation TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism

glutamine asparate palmitate malate acetate fumarate fumarate fumarate fumarate a-ketoglutarate a-ketoglutarate a-ketoglutarate a-ketoglutarate tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine tyrosine fumarate phenylalanine

5.41E+06 -3.71E+06 -7.76E+05 1.72E+05 -2.09E+04 1.43E+06

3.63E+05 8.13E+06 4.21E+06 6.31E+04 4.78E+04 2.43E+05

5.77E+06 4.42E+06 3.43E+06 2.35E+05 2.69E+04 1.67E+06

9.19E+04

14112.359

106009.21

1.12E+06

7.58E+05

1.88E+06

159

1.43E+06 4.06E+05

2.43E+05 3.10E+05

1.67E+06 7.17E+05

R03660 R00689 R00694 R00698 R00699 R02382 R02383 RE2124 RE2781 R00261 R01648 R01986

Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine

phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine tyramine tyramine tyramine tyramine 4aminobutanoate 4aminobutanoate 4aminobutanoate 4aminobutanoate 4aminobutanoate a-ketoglutarate a-ketoglutarate a-ketoglutarate a-ketoglutarate a-ketoglutarate a-ketoglutarate asparate & glutamine -3.71E+06 8.13E+06 4.42E+06 -2.67E+04 31646.636 4927.352

36077.002

37757.204

73834.206

160

R02549 R01987 R00243 R01248 R00248 R00270 R00258 R03534 R00578

91896.851

14112.359

106009.21

R03421 R00526 R05577 R03647 R00357 R00488 R00355 R01954 R01251

R00489 R00253 R00256 R03652 R00667 R00669 R01398 R00135 R03661 R01252

Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and

asparate asparate asparate asparate asparate asparate asparate asparate asparate asparate glutamine glutamine glutamine ornithine ornithine ornithine proline proline proline 50577.225 26032.511 76609.736 -113645.303 302159.156 188513.853 5.41E+06 3.63E+05 5.77E+06

161

asparagine RE1537 R01920 R01989 Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine putrescine putrescine putrescine & 4aminobutanoate putrescine putrescine putrescine putrescine fumarate glycine glycine urea asparagine asparagine asparagine leucine leucine valine valine 1.62E+06 2.60E+06 4.22E+06 2.75E+06 2.95E+06 5.69E+06 7.46E+05 4.03E+03 75710.785 8265.44 821237.285 12295.035 1.43E+06 4.49E+06 2.43E+05 730562.804 1.67E+06 5222233.54

R01151 R01154 R01157 R00670

R01086 R00497 R04951 R00551 R03648 RE2032 RE2644 R03657 R01090 R03665 R01214

Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation

162

R01360 R00257

Valine, leucine and isoleucine degradation Vitamin B3 (nicotinate and nicotinamide) metabolism

acetoacetate glutamine

3.40E+03 5.41E+06

6727.541 3.63E+05

10130.481 5.77E+06

163

Appendix F List of Pathways Identified by Metscape Analysis from 7 mM glucose to 17 mM Glucose with Reactions, Seeds Involved, Direction of Change and Peak Areas

164

Average Peak Area Direction of Change

Reaction R00768 R07041 R01593 RE3651 R07046 R07051 RE3287 R01317 RE3650 RE3288

Pathway Aminosugars metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism Arachidonic acid metabolism

seed glutamine arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate glycerol

17 mM - 7 mM -1.15E+07 1.34E+04

7 mM 1.61E+07 3.15E+04

17 mM 4.59E+06 4.49E+04

165

R07048 R07050 RE3646 RE3042 RE3649 RE3286 RE3654 RE3647 RE3653 R01596 RE3648 RE3289 R07054 R07052 RE2563

1.41E+04

30179.373

44244.974

RE2754 RE2638 R03718 RE2649 RE2650 R01357 R00410 R01361 R01176 RE1575 RE1514 R00866 R00875 RE2783 R00867 R01326 R01105 R01101 R01678 R01095 R01329 R02926 R03634 R05549 R01092 R01104

Arachidonic acid metabolism Bile acid biosynthesis Bile acid biosynthesis Bile acid biosynthesis Biopterin metabolism Butanoate metabolism Butanoate metabolism Butanoate metabolism Butanoate metabolism De novo fatty acid biosynthesis De novo fatty acid biosynthesis Di-unsaturated fatty acid beta-oxidation Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Fructose and mannose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism Galactose metabolism

glycerol glycine glycine propanoate tyrosine acetoacetate acetoacetate acetoacetate butanoate stearate lauric acid linoleate d-fructose d-fructose d-fructose d-fructose d-mannose d-galactose d-galactose d-galactose d-galactose d-galactose & d-mannose d-galactose d-galactose d-galactose d-galactose d-galactose -1.96E+04 2.84E+04 8.84E+03 -1.96E+04 -4.03E+06 2.84E+04 8.11E+06 8.84E+03 4.08E+06 1.86E+06 9.72E+05 3.47E+03 3.35E+06 9.05E+05 6655.777 5.21E+06 1.88E+06 10130.481 -7.43E+04 1.70E+05 9.61E+04

-7.91E+05 1.02E+03 7.52E+03 -5.73E+04 9.35E+03

1.62E+06 15157.211 4.26E+03 2.39E+05 10660.345

8.29E+05 16176.875 1.18E+04 1.81E+05 20009.769

166

R01093 R01103 R01100 RE3301 RE3298 RE3299 R01351 R01352 R01350 RE3507 R01041 R03616 R00847 R01036 R04452 R03038 RE2642 RE1473 RE2031 R00369 RE2429 R03654 R00366 RE2111 R06171 RE2117

Galactose metabolism Galactose metabolism Galactose metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycerophospholipid metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism

d-galactose d-galactose & sucrose d-galactose serine serine serine glycerol glycerol glycerol glycerol glycerol glycerol glycerol glycerol acetate alanine alanine alanine alanine alanine glycine glycine glycine & glyoxylate glycine glycine glycine -7.43E+04 1.70E+05 9.61E+04 -5.98E+03 -9.79E+04 4.09E+04 1.59E+05 3.49E+04 6.07E+04 1.41E+04 30179.373 44244.974 9.65E+05 1.23E+06 2.20E+06 -3.74E+03 1.34E+04 9.63E+03

167

R00367 RE2427 R00945 RE2428 R00751 R00610 R01221 R00565 R00611 R03425 R00371 R00465 R00475 R00717 RO2821 R00704 R03662 R00582 R00585 R00220 R03663 R00317 R00711 R00316 R00235 R00710

Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycine, serine, alanine and threonine metabolism Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis

glycine glycine glycine & serine glycine glycine & threonine glycine glycine glycine glycine glycine glycine glyoxylate glyoxylate glyoxylate methionine pyruvate serine serine serine serine threonine acetate acetate acetate acetate acetate -5.98E+03 4.09E+04 3.49E+04 1.51E+05 185057.809 9.65E+05 118295.916 240399.287 1.23E+06 269211.712 425457.096 2.20E+06 7.38E+04 6.98E+04 1.44E+05 2.32E+05 5.35E+05 7.67E+05

168

R00227 R00196 R00006 R00014 R00200 R00209 R00210 R00214 R00216 R00344 R00703 R01699 RE2717 R01281 RE3632 R01595 RE0596 RE1821 RE3417 RE3420 R07063 RE3409 R07056 R07057 R07055 R03626 R07064 RE3411

Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycolysis and Gluconeogenesis Glycosphingolipid metabolism Glycosphingolipid metabolism Histidine metabolism Leukotriene metabolism Leukotriene metabolism Leukotriene metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism Linoleate metabolism

acetate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate pyruvate d-galactose serine glycine arachidonate glycine glycine azelaic acid azelaic acid linoleate linoleate linoleate linoleate linoleate linoleate linoleate nonanoate 1.99E+04 62086.16 82035.258 -5.73E+04 2.39E+05 1.81E+05 1.89E+04 5.48E+04 7.37E+04 -4.03E+06 9.65E+05 -7.43E+04 1.34E+04 -7.43E+04 8.11E+06 1.23E+06 1.70E+05 3.15E+04 1.70E+05 4.08E+06 2.20E+06 9.61E+04 4.49E+04 9.61E+04 185057.809 240399.287 425457.096

169

R02201 R02205 R02203 R02204 R03284 R00652 R00892 RE2811 R01001 RE2223 R00893 R00782 R03650 R00896 R00946 RE2030 R00648 R00177 R03659 RE2640 R03105 R00590 R01290 R00891 R01289 RE0324

Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Lysine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Methionine and cysteine metabolism Omega-6 fatty acid metabolism

pipecolate pipecolate pipecolate pipecolate glycine glycine & methionine cysteine cysteine cysteine cysteine cysteine cysteine cysteine cysteine methionine methionine methionine methionine methionine methionine pyruvate serine serine serine & cysteine serine arachidonate

-226936.216

574259.195

347322.979

-7.43E+04 -7.43E+04 1.51E+05 1.27E+05

1.70E+05 1.70E+05 118295.916 209254.529

9.61E+04 9.61E+04 269211.712 336439.947

170

185057.809 9.65E+05

240399.287 1.23E+06

425457.096 2.20E+06

1.34E+04

3.15E+04

4.49E+04

RE0124 R01641 R06590 R01057 R01051 R01054 R01056 R01049 R01520 R01521 R01194 R03331 R00830 R00831 R01354 R01590 RE3449 RE3452 RE3455 RE1978 RE3458 RE2067 RE2068 R01231 R01072 R01083 R01135

Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Pentose phosphate pathway Phosphatidylinositol phosphate metabolism Phosphatidylinositol phosphate metabolism Porphyrin metabolism Porphyrin metabolism Propanoate metabolism Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Prostaglandin formation from arachidonate Purine metabolism Purine metabolism Purine metabolism Purine metabolism

d-gluconic acid ribose-5-phospate ribose-5-phospate ribose-5-phospate ribose-5-phospate ribose-5-phospate ribose-5-phospate ribose-5-phospate d-glucono-1,5lactone d-glucono-1,5lactone d-galactose glycerol glycine glycine propanoate arachidonate arachidonate arachidonate arachidonate arachidonate arachidonate glycerol glycerol glutamine glutamine fumarate aspartate

3.60E+03 -1.25E+04

4546.984 8.04E+04

8144.155 6.79E+04

2.43E+03

6.22E+03

8.65E+03

-4.03E+06 1.41E+04 -7.43E+04 1.86E+06 1.34E+04

8.11E+06 30179.373 1.70E+05 3.35E+06 3.15E+04

4.08E+06 44244.974 9.61E+04 5.21E+06 4.49E+04

171

1.41E+04 -1.15E+07 6.19E+05 -3.14E+06

30179.373 1.61E+07 1.05E+06 5.78E+06

44244.974 4.59E+06 1.67E+06 2.64E+06

R00575 R01397 R02327 R01880 R00963 R00967 R01878 R00970 R01549 R01876 R00964 R02097 R00968 R02332 R02164 R01082 R00408 R00412 R00268 R00709 RE0361 R00621 R00362 R01364 RE0941 R02918 RE3062

Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism Pyrimidine metabolism TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle TCA cycle Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism

glutamine aspartate uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine uridine fumarate fumarate & malate fumarate fumarate a-ketoglutarate a-ketoglutarate a-ketoglutarate a-ketoglutarate acetate glutamine glutamine glutamine glutamine

-1.15E+07 -3.14E+06 -1.70E+04

1.61E+07 5.78E+06 5.54E+04

4.59E+06 2.64E+06 3.84E+04

172

6.19E+05

1.05E+06

1.67E+06

5.10E+04

5.93E+04

1.10E+05

-5.98E+03 -1.15E+07

4.09E+04 1.61E+07

3.49E+04 4.59E+06

R00731 R01815 R00734 R02078 RE0026 R00729 R03539 R03660 R00694 R00698 R00699 R00692 RE1465 R00689 R01795 R00736

Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism

glutamine glutamine glutamine glutamine glutamine glutamine glutamine phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine phenylalanine tyrosine 3,4dihydroxyphenoleth yleneglycol 3,4dihydroxyphenoleth yleneglycol cysteine pyruvate 5-oxoproline 5-oxoproline 5-oxoproline a-ketoglutarate 5.10E+04 5.93E+04 1.10E+05 9.72E+05 9.05E+05 1.88E+06 2.78E+05 3.31E+05 6.09E+05

173

R04880

Tyrosine metabolism

1.55E+04

20282.07

35816.138

R04881 RE1915 RE1919 R02743 R03749 R00251 R03534

Tyrosine metabolism Tyrosine metabolism Tyrosine metabolism Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine

1.27E+05 185057.809 4.04E+05

209254.529 240399.287 1.25E+05

336439.947 425457.096 5.29E+05

R00243 R00355

Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate,

a-ketoglutarate a-ketoglutarate & aspartate

R00270 R00258

a-ketoglutarate a-ketoglutarate & alanine -9.79E+04 1.59E+05 6.07E+04

R00248 R05577 R00488 R01954 R00526 R03421 R00489 R03647 R00578

a-ketoglutarate aspartate aspartate aspartate aspartate aspartate aspartate aspartate aspartate & glutamine -3.14E+06 5.78E+06 2.64E+06

174

R00357 R01086 R00253 R00256

aspartate fumarate glutamine glutamine 6.19E+05 -1.15E+07 1.05E+06 1.61E+07 1.67E+06 4.59E+06

aspartate and asparagine R03652 R00497 R04951 R00894 R00899 R03648 RE2032 RE2644 R03657 R01090 R01214 R03665 R01360 R00257 R04233 R04230 Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Urea cycle and metabolism of arginine, proline, glutamate, aspartate and asparagine Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Valine, leucine and isoleucine degradation Vitamin B3 (nicotinate and nicotinamide) metabolism Vitamin B5 - CoA biosynthesis from pantothenate Vitamin B5 - CoA biosynthesis from pantothenate glutamine glycine glycine cysteine cysteine asparagine asparagine asparagine leucine leucine valine valine acetoacetate glutamine cysteine cysteine 3.47E+03 -1.15E+07 1.27E+05 6655.777 1.61E+07 209254.529 10130.481 4.59E+06 336439.947 1.77E+04 2.50E+04 4.27E+04 2.36E+06 3.33E+06 5.69E+06 -6537.452 21844.953 15307.501 1.27E+05 209254.529 336439.947 -7.43E+04 1.70E+05 9.61E+04

175

Appendix G Location of Raw and Processed Data Files for Chapters 2 & 3 All raw data files for Chapters 2 and 3 are backed up on the lab hard drive in appropriately marked folders and remain on the Pegasus Computer. Additionally, all of the raw data files for Chapter 2 as well as archived versions of the data processing methods and references can be found on the hard drive provided for Bob. Archived information can be restored directly into ChromatTOF and used to process data without having to rewrite the methods if desired.
Processed Data Files

For additional ChromaTOF user information see the manual which can be accessed as a PDF by opening ChromaTOF as directed below and clicking help followed by ChromaTOF help.
Chapter 2

All processed chromatograms used for the metabolite profiling analysis are located on the Pegasus Workstation (Drive E) computer. To start the software, double click the gray Pegasus icon and click login (no password needed). On the left hand side, click Acquired Samples. All of the processed chromatograms are in the folder labeled 3 to 7 to 17. An example sample name is as follows: 7mM A (5mVIIa):3 where 7 is the concentration of glucose, A is the plate identification, 5 is the page number in notebook mVII (m distinguishes the metabolomics notebooks from the sapropel notebooks) and a identifies the replicate on that page. The number after the colon is 176

added automatically by the software to distinguish samples with the same name. In this case, that would occur, for example, if the autosampler were to misfire and the run to be restarted with the same name. Data was processed with both a 0.1 and 0.2 second peak width for the second dimension. Shown below is the method for the 0.2 second peak width and the box that should be changed to 0.1 for the 0.1 second method is highlighted in orange.

177

178

The reference 3 to 7 0.2 referred to above is a previously run chromatogram that has all of the target peaks identified with *. After processing, target analytes will also have * next to their identifications and the peak areas will all be calculated using the same m/z ratio. References must be processed using the exact same processing method (minus the reference) as the chromatogram for which they are added, therefore, there is a reference for the 0.2 second and the 0.1 second data processing methods. These can be found under

179

References on the left side of the screen when ChromaTOF is open. The references used in this work were calculated using a first dimension retention time deviation of 7 seconds (one modulation period) and a second dimension retention time deviation of 0.2 seconds. The quant mass (or the m/z that the peak area is calculated at) of an analyte that was processed using a reference cannot be changed. Therefore, succinate and glucose are not included in the references. This way, the quant mass can be manually changed to calculate the area of both the labeled and unlabeled analyte. For more information on creating a reference see Building a Reference on page 5-125 of the manual. Alanine, valine, proline, glycine, fumarate, threonine, malate, aspartate, phenylalanine, 3-phosphoglycerate, ornithine, tyrosine, ribose-5-phosphate, glucose-6phosphate, succinate and methionine data were taken from the 0.2 second processing results. Lactate, pyruvate, leucine, isoleucine, serine, glutamate, lysine and glucose were taken from the 0.1 second data processing. The peak area, average peak areas, standard deviations, relative standard deviations and results of Q-tests can be found in the excel file 3,7,17 july located in the gly-tca data folder on the desktop. All cells highlighted in dark red were deleted and if average values were not calculated, the plate was not included in the final data set. All of the files created for Chapter 2 are also included in this folder and the prism files used to complete one-way ANOVE analysis are located in the Chapter 2 subfolder.
Fisher Ratio Data

In order to calculate a Fisher Ratio, data sets must be placed into groups within the software. Because of the way the software works, once a file has been added to a 180

group, it cannot be added to a second group. In order to get around this, each data file was exported and then imported with a different number added to the end of the file name and then reprocessed. All files used for the Fisher Ratio analysis were only processed using the 0.2 second data processing method. The reprocessed files can be found in the Fisher 3 to 7, Fisher 3 to 17, Fisher 7 to 17, and Fisher 3 to 3 folders located under Acquired Samples. The actual Fisher Ratio results can be located under Statistical Compare and are labeled as 3 to 7 (july), 3 to 17 (july), 7 to 17 (july), and 3 to 3 (july). The gas chromatograph, mass spec and autosampler methods used to run the samples discussed in Chapter 2 follow.

181

Gas Chromatograph Method:

182

183

184

Mass Spectral Method:

185

Autosampler Method:

186

Chapter 3

Processed data files for the samples discussed in Chapter 3 can also be found on the Pegasus Workstation Computer (Drive C). The files are located in a secondary database and the software must be opened differently than previously described to access the Chapter 3 files. Double click on the gray Pegasus icon. Click change new ok yes login. Again, no password is needed. The processed chromatograms are located in the FAMEs Paper folder under Acquired Samples. The gas chromatograph, mass spec, autosampler and data processing methods used follow.

187

Gas Chromatograph Method:

188

189

190

Mass Spectral Method:

191

Autosampler Method:

192

Data Processing Method:

193

194

195